@lobehub/chat 1.68.7 → 1.68.8
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +25 -0
- package/changelog/v1.json +9 -0
- package/package.json +1 -1
- package/src/database/server/models/aiModel.ts +1 -0
- package/src/libs/agent-runtime/openrouter/__snapshots__/index.test.ts.snap +898 -48
- package/src/libs/agent-runtime/openrouter/index.ts +24 -8
- package/src/libs/agent-runtime/openrouter/type.ts +11 -0
- package/src/types/llm.ts +1 -1
|
@@ -14,7 +14,12 @@ _These are free, rate-limited endpoints for [Reflection 70B](/models/mattshumer/
|
|
|
14
14
|
"functionCall": false,
|
|
15
15
|
"id": "mattshumer/reflection-70b:free",
|
|
16
16
|
"maxTokens": 4096,
|
|
17
|
+
"pricing": {
|
|
18
|
+
"input": 0,
|
|
19
|
+
"output": 0,
|
|
20
|
+
},
|
|
17
21
|
"reasoning": false,
|
|
22
|
+
"releasedAt": "2024-09-06",
|
|
18
23
|
"vision": false,
|
|
19
24
|
},
|
|
20
25
|
{
|
|
@@ -27,7 +32,12 @@ The model was trained on synthetic data.",
|
|
|
27
32
|
"functionCall": false,
|
|
28
33
|
"id": "mattshumer/reflection-70b",
|
|
29
34
|
"maxTokens": undefined,
|
|
35
|
+
"pricing": {
|
|
36
|
+
"input": 0.35,
|
|
37
|
+
"output": 0.4,
|
|
38
|
+
},
|
|
30
39
|
"reasoning": false,
|
|
40
|
+
"releasedAt": "2024-09-06",
|
|
31
41
|
"vision": false,
|
|
32
42
|
},
|
|
33
43
|
{
|
|
@@ -39,10 +49,15 @@ Read the launch post [here](https://txt.cohere.com/command-r/).
|
|
|
39
49
|
Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).",
|
|
40
50
|
"displayName": "Cohere: Command R (03-2024)",
|
|
41
51
|
"enabled": false,
|
|
42
|
-
"functionCall":
|
|
52
|
+
"functionCall": true,
|
|
43
53
|
"id": "cohere/command-r-03-2024",
|
|
44
54
|
"maxTokens": 4000,
|
|
55
|
+
"pricing": {
|
|
56
|
+
"input": 0.5,
|
|
57
|
+
"output": 1.5,
|
|
58
|
+
},
|
|
45
59
|
"reasoning": false,
|
|
60
|
+
"releasedAt": "2024-08-31",
|
|
46
61
|
"vision": false,
|
|
47
62
|
},
|
|
48
63
|
{
|
|
@@ -54,10 +69,15 @@ It offers multilingual support for ten key languages to facilitate global busine
|
|
|
54
69
|
Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).",
|
|
55
70
|
"displayName": "Cohere: Command R+ (04-2024)",
|
|
56
71
|
"enabled": false,
|
|
57
|
-
"functionCall":
|
|
72
|
+
"functionCall": true,
|
|
58
73
|
"id": "cohere/command-r-plus-04-2024",
|
|
59
74
|
"maxTokens": 4000,
|
|
75
|
+
"pricing": {
|
|
76
|
+
"input": 3,
|
|
77
|
+
"output": 15,
|
|
78
|
+
},
|
|
60
79
|
"reasoning": false,
|
|
80
|
+
"releasedAt": "2024-08-31",
|
|
61
81
|
"vision": false,
|
|
62
82
|
},
|
|
63
83
|
{
|
|
@@ -69,10 +89,15 @@ Read the launch post [here](https://docs.cohere.com/changelog/command-gets-refre
|
|
|
69
89
|
Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).",
|
|
70
90
|
"displayName": "Cohere: Command R+ (08-2024)",
|
|
71
91
|
"enabled": false,
|
|
72
|
-
"functionCall":
|
|
92
|
+
"functionCall": true,
|
|
73
93
|
"id": "cohere/command-r-plus-08-2024",
|
|
74
94
|
"maxTokens": 4000,
|
|
95
|
+
"pricing": {
|
|
96
|
+
"input": 2.5,
|
|
97
|
+
"output": 10,
|
|
98
|
+
},
|
|
75
99
|
"reasoning": false,
|
|
100
|
+
"releasedAt": "2024-08-30",
|
|
76
101
|
"vision": false,
|
|
77
102
|
},
|
|
78
103
|
{
|
|
@@ -84,10 +109,15 @@ Read the launch post [here](https://docs.cohere.com/changelog/command-gets-refre
|
|
|
84
109
|
Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).",
|
|
85
110
|
"displayName": "Cohere: Command R (08-2024)",
|
|
86
111
|
"enabled": false,
|
|
87
|
-
"functionCall":
|
|
112
|
+
"functionCall": true,
|
|
88
113
|
"id": "cohere/command-r-08-2024",
|
|
89
114
|
"maxTokens": 4000,
|
|
115
|
+
"pricing": {
|
|
116
|
+
"input": 0.15,
|
|
117
|
+
"output": 0.6,
|
|
118
|
+
},
|
|
90
119
|
"reasoning": false,
|
|
120
|
+
"releasedAt": "2024-08-30",
|
|
91
121
|
"vision": false,
|
|
92
122
|
},
|
|
93
123
|
{
|
|
@@ -104,7 +134,12 @@ Note: This model is experimental and not suited for production use-cases. It may
|
|
|
104
134
|
"functionCall": false,
|
|
105
135
|
"id": "google/gemini-flash-8b-1.5-exp",
|
|
106
136
|
"maxTokens": 32768,
|
|
137
|
+
"pricing": {
|
|
138
|
+
"input": 0,
|
|
139
|
+
"output": 0,
|
|
140
|
+
},
|
|
107
141
|
"reasoning": false,
|
|
142
|
+
"releasedAt": "2024-08-28",
|
|
108
143
|
"vision": true,
|
|
109
144
|
},
|
|
110
145
|
{
|
|
@@ -121,7 +156,12 @@ Note: This model is experimental and not suited for production use-cases. It may
|
|
|
121
156
|
"functionCall": false,
|
|
122
157
|
"id": "google/gemini-flash-1.5-exp",
|
|
123
158
|
"maxTokens": 32768,
|
|
159
|
+
"pricing": {
|
|
160
|
+
"input": 0,
|
|
161
|
+
"output": 0,
|
|
162
|
+
},
|
|
124
163
|
"reasoning": false,
|
|
164
|
+
"releasedAt": "2024-08-28",
|
|
125
165
|
"vision": true,
|
|
126
166
|
},
|
|
127
167
|
{
|
|
@@ -132,7 +172,12 @@ Note: This model is experimental and not suited for production use-cases. It may
|
|
|
132
172
|
"functionCall": false,
|
|
133
173
|
"id": "sao10k/l3.1-euryale-70b",
|
|
134
174
|
"maxTokens": undefined,
|
|
175
|
+
"pricing": {
|
|
176
|
+
"input": 1.5,
|
|
177
|
+
"output": 1.5,
|
|
178
|
+
},
|
|
135
179
|
"reasoning": false,
|
|
180
|
+
"releasedAt": "2024-08-28",
|
|
136
181
|
"vision": false,
|
|
137
182
|
},
|
|
138
183
|
{
|
|
@@ -146,10 +191,15 @@ Built on a novel SSM-Transformer architecture, it outperforms larger models like
|
|
|
146
191
|
Read their [announcement](https://www.ai21.com/blog/announcing-jamba-model-family) to learn more.",
|
|
147
192
|
"displayName": "AI21: Jamba 1.5 Large",
|
|
148
193
|
"enabled": false,
|
|
149
|
-
"functionCall":
|
|
194
|
+
"functionCall": true,
|
|
150
195
|
"id": "ai21/jamba-1-5-large",
|
|
151
196
|
"maxTokens": 4096,
|
|
197
|
+
"pricing": {
|
|
198
|
+
"input": 2,
|
|
199
|
+
"output": 8,
|
|
200
|
+
},
|
|
152
201
|
"reasoning": false,
|
|
202
|
+
"releasedAt": "2024-08-23",
|
|
153
203
|
"vision": false,
|
|
154
204
|
},
|
|
155
205
|
{
|
|
@@ -163,10 +213,15 @@ This model uses less computer memory and works faster with longer texts than pre
|
|
|
163
213
|
Read their [announcement](https://www.ai21.com/blog/announcing-jamba-model-family) to learn more.",
|
|
164
214
|
"displayName": "AI21: Jamba 1.5 Mini",
|
|
165
215
|
"enabled": false,
|
|
166
|
-
"functionCall":
|
|
216
|
+
"functionCall": true,
|
|
167
217
|
"id": "ai21/jamba-1-5-mini",
|
|
168
218
|
"maxTokens": 4096,
|
|
219
|
+
"pricing": {
|
|
220
|
+
"input": 0.2,
|
|
221
|
+
"output": 0.4,
|
|
222
|
+
},
|
|
169
223
|
"reasoning": false,
|
|
224
|
+
"releasedAt": "2024-08-23",
|
|
170
225
|
"vision": false,
|
|
171
226
|
},
|
|
172
227
|
{
|
|
@@ -176,10 +231,15 @@ Read their [announcement](https://www.ai21.com/blog/announcing-jamba-model-famil
|
|
|
176
231
|
The models underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures. When assessed against benchmarks that test common sense, language understanding, math, code, long context and logical reasoning, Phi-3.5 models showcased robust and state-of-the-art performance among models with less than 13 billion parameters.",
|
|
177
232
|
"displayName": "Phi-3.5 Mini 128K Instruct",
|
|
178
233
|
"enabled": false,
|
|
179
|
-
"functionCall":
|
|
234
|
+
"functionCall": true,
|
|
180
235
|
"id": "microsoft/phi-3.5-mini-128k-instruct",
|
|
181
236
|
"maxTokens": undefined,
|
|
237
|
+
"pricing": {
|
|
238
|
+
"input": 0.1,
|
|
239
|
+
"output": 0.1,
|
|
240
|
+
},
|
|
182
241
|
"reasoning": false,
|
|
242
|
+
"releasedAt": "2024-08-21",
|
|
183
243
|
"vision": false,
|
|
184
244
|
},
|
|
185
245
|
{
|
|
@@ -194,7 +254,12 @@ The Hermes 3 series builds and expands on the Hermes 2 set of capabilities, incl
|
|
|
194
254
|
"functionCall": true,
|
|
195
255
|
"id": "nousresearch/hermes-3-llama-3.1-70b",
|
|
196
256
|
"maxTokens": undefined,
|
|
257
|
+
"pricing": {
|
|
258
|
+
"input": 0.4,
|
|
259
|
+
"output": 0.4,
|
|
260
|
+
},
|
|
197
261
|
"reasoning": false,
|
|
262
|
+
"releasedAt": "2024-08-18",
|
|
198
263
|
"vision": false,
|
|
199
264
|
},
|
|
200
265
|
{
|
|
@@ -211,7 +276,12 @@ Hermes 3 is competitive, if not superior, to Llama-3.1 Instruct models at genera
|
|
|
211
276
|
"functionCall": true,
|
|
212
277
|
"id": "nousresearch/hermes-3-llama-3.1-405b",
|
|
213
278
|
"maxTokens": undefined,
|
|
279
|
+
"pricing": {
|
|
280
|
+
"input": 0,
|
|
281
|
+
"output": 0,
|
|
282
|
+
},
|
|
214
283
|
"reasoning": false,
|
|
284
|
+
"releasedAt": "2024-08-16",
|
|
215
285
|
"vision": false,
|
|
216
286
|
},
|
|
217
287
|
{
|
|
@@ -230,7 +300,12 @@ _These are extended-context endpoints for [Hermes 3 405B Instruct](/models/nousr
|
|
|
230
300
|
"functionCall": true,
|
|
231
301
|
"id": "nousresearch/hermes-3-llama-3.1-405b:extended",
|
|
232
302
|
"maxTokens": undefined,
|
|
303
|
+
"pricing": {
|
|
304
|
+
"input": 0,
|
|
305
|
+
"output": 0,
|
|
306
|
+
},
|
|
233
307
|
"reasoning": false,
|
|
308
|
+
"releasedAt": "2024-08-16",
|
|
234
309
|
"vision": false,
|
|
235
310
|
},
|
|
236
311
|
{
|
|
@@ -241,7 +316,12 @@ _These are extended-context endpoints for [Hermes 3 405B Instruct](/models/nousr
|
|
|
241
316
|
"functionCall": false,
|
|
242
317
|
"id": "perplexity/llama-3.1-sonar-huge-128k-online",
|
|
243
318
|
"maxTokens": undefined,
|
|
319
|
+
"pricing": {
|
|
320
|
+
"input": 5,
|
|
321
|
+
"output": 5,
|
|
322
|
+
},
|
|
244
323
|
"reasoning": false,
|
|
324
|
+
"releasedAt": "2024-08-14",
|
|
245
325
|
"vision": false,
|
|
246
326
|
},
|
|
247
327
|
{
|
|
@@ -254,8 +334,13 @@ Note: This model is experimental and not suited for production use-cases. It may
|
|
|
254
334
|
"functionCall": false,
|
|
255
335
|
"id": "openai/chatgpt-4o-latest",
|
|
256
336
|
"maxTokens": 16384,
|
|
337
|
+
"pricing": {
|
|
338
|
+
"input": 5,
|
|
339
|
+
"output": 15,
|
|
340
|
+
},
|
|
257
341
|
"reasoning": false,
|
|
258
|
-
"
|
|
342
|
+
"releasedAt": "2024-08-14",
|
|
343
|
+
"vision": true,
|
|
259
344
|
},
|
|
260
345
|
{
|
|
261
346
|
"contextWindowTokens": 8192,
|
|
@@ -269,7 +354,12 @@ For best results, use with Llama 3 Instruct context template, temperature 1.4, a
|
|
|
269
354
|
"functionCall": false,
|
|
270
355
|
"id": "sao10k/l3-lunaris-8b",
|
|
271
356
|
"maxTokens": undefined,
|
|
357
|
+
"pricing": {
|
|
358
|
+
"input": 2,
|
|
359
|
+
"output": 2,
|
|
360
|
+
},
|
|
272
361
|
"reasoning": false,
|
|
362
|
+
"releasedAt": "2024-08-13",
|
|
273
363
|
"vision": false,
|
|
274
364
|
},
|
|
275
365
|
{
|
|
@@ -282,7 +372,12 @@ Although more similar to Magnum overall, the model remains very creative, with a
|
|
|
282
372
|
"functionCall": false,
|
|
283
373
|
"id": "aetherwiing/mn-starcannon-12b",
|
|
284
374
|
"maxTokens": undefined,
|
|
375
|
+
"pricing": {
|
|
376
|
+
"input": 2,
|
|
377
|
+
"output": 2,
|
|
378
|
+
},
|
|
285
379
|
"reasoning": false,
|
|
380
|
+
"releasedAt": "2024-08-13",
|
|
286
381
|
"vision": false,
|
|
287
382
|
},
|
|
288
383
|
{
|
|
@@ -294,11 +389,16 @@ GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and im
|
|
|
294
389
|
For benchmarking against other models, it was briefly called ["im-also-a-good-gpt2-chatbot"](https://twitter.com/LiamFedus/status/1790064963966370209)",
|
|
295
390
|
"displayName": "OpenAI: GPT-4o (2024-08-06)",
|
|
296
391
|
"enabled": false,
|
|
297
|
-
"functionCall":
|
|
392
|
+
"functionCall": true,
|
|
298
393
|
"id": "openai/gpt-4o-2024-08-06",
|
|
299
394
|
"maxTokens": 16384,
|
|
395
|
+
"pricing": {
|
|
396
|
+
"input": 2.5,
|
|
397
|
+
"output": 10,
|
|
398
|
+
},
|
|
300
399
|
"reasoning": false,
|
|
301
|
-
"
|
|
400
|
+
"releasedAt": "2024-08-06",
|
|
401
|
+
"vision": true,
|
|
302
402
|
},
|
|
303
403
|
{
|
|
304
404
|
"contextWindowTokens": 131072,
|
|
@@ -312,7 +412,12 @@ To read more about the model release, [click here](https://ai.meta.com/blog/meta
|
|
|
312
412
|
"functionCall": false,
|
|
313
413
|
"id": "meta-llama/llama-3.1-405b",
|
|
314
414
|
"maxTokens": undefined,
|
|
415
|
+
"pricing": {
|
|
416
|
+
"input": 2,
|
|
417
|
+
"output": 2,
|
|
418
|
+
},
|
|
315
419
|
"reasoning": false,
|
|
420
|
+
"releasedAt": "2024-08-02",
|
|
316
421
|
"vision": false,
|
|
317
422
|
},
|
|
318
423
|
{
|
|
@@ -327,7 +432,12 @@ Check out the model's [HuggingFace page](https://huggingface.co/nothingiisreal/M
|
|
|
327
432
|
"functionCall": false,
|
|
328
433
|
"id": "nothingiisreal/mn-celeste-12b",
|
|
329
434
|
"maxTokens": undefined,
|
|
435
|
+
"pricing": {
|
|
436
|
+
"input": 1.5,
|
|
437
|
+
"output": 1.5,
|
|
438
|
+
},
|
|
330
439
|
"reasoning": false,
|
|
440
|
+
"releasedAt": "2024-08-02",
|
|
331
441
|
"vision": false,
|
|
332
442
|
},
|
|
333
443
|
{
|
|
@@ -344,7 +454,12 @@ Note: This model is experimental and not suited for production use-cases. It may
|
|
|
344
454
|
"functionCall": false,
|
|
345
455
|
"id": "google/gemini-pro-1.5-exp",
|
|
346
456
|
"maxTokens": 32768,
|
|
457
|
+
"pricing": {
|
|
458
|
+
"input": 0,
|
|
459
|
+
"output": 0,
|
|
460
|
+
},
|
|
347
461
|
"reasoning": false,
|
|
462
|
+
"releasedAt": "2024-08-01",
|
|
348
463
|
"vision": true,
|
|
349
464
|
},
|
|
350
465
|
{
|
|
@@ -357,7 +472,12 @@ This is the online version of the [offline chat model](/models/perplexity/llama-
|
|
|
357
472
|
"functionCall": false,
|
|
358
473
|
"id": "perplexity/llama-3.1-sonar-large-128k-online",
|
|
359
474
|
"maxTokens": undefined,
|
|
475
|
+
"pricing": {
|
|
476
|
+
"input": 1,
|
|
477
|
+
"output": 1,
|
|
478
|
+
},
|
|
360
479
|
"reasoning": false,
|
|
480
|
+
"releasedAt": "2024-08-01",
|
|
361
481
|
"vision": false,
|
|
362
482
|
},
|
|
363
483
|
{
|
|
@@ -370,7 +490,12 @@ This is a normal offline LLM, but the [online version](/models/perplexity/llama-
|
|
|
370
490
|
"functionCall": false,
|
|
371
491
|
"id": "perplexity/llama-3.1-sonar-large-128k-chat",
|
|
372
492
|
"maxTokens": undefined,
|
|
493
|
+
"pricing": {
|
|
494
|
+
"input": 1,
|
|
495
|
+
"output": 1,
|
|
496
|
+
},
|
|
373
497
|
"reasoning": false,
|
|
498
|
+
"releasedAt": "2024-08-01",
|
|
374
499
|
"vision": false,
|
|
375
500
|
},
|
|
376
501
|
{
|
|
@@ -383,7 +508,12 @@ This is the online version of the [offline chat model](/models/perplexity/llama-
|
|
|
383
508
|
"functionCall": false,
|
|
384
509
|
"id": "perplexity/llama-3.1-sonar-small-128k-online",
|
|
385
510
|
"maxTokens": undefined,
|
|
511
|
+
"pricing": {
|
|
512
|
+
"input": 0.2,
|
|
513
|
+
"output": 0.2,
|
|
514
|
+
},
|
|
386
515
|
"reasoning": false,
|
|
516
|
+
"releasedAt": "2024-08-01",
|
|
387
517
|
"vision": false,
|
|
388
518
|
},
|
|
389
519
|
{
|
|
@@ -396,7 +526,12 @@ This is a normal offline LLM, but the [online version](/models/perplexity/llama-
|
|
|
396
526
|
"functionCall": false,
|
|
397
527
|
"id": "perplexity/llama-3.1-sonar-small-128k-chat",
|
|
398
528
|
"maxTokens": undefined,
|
|
529
|
+
"pricing": {
|
|
530
|
+
"input": 0.2,
|
|
531
|
+
"output": 0.2,
|
|
532
|
+
},
|
|
399
533
|
"reasoning": false,
|
|
534
|
+
"releasedAt": "2024-08-01",
|
|
400
535
|
"vision": false,
|
|
401
536
|
},
|
|
402
537
|
{
|
|
@@ -408,10 +543,15 @@ It has demonstrated strong performance compared to leading closed-source models
|
|
|
408
543
|
To read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).",
|
|
409
544
|
"displayName": "Meta: Llama 3.1 70B Instruct",
|
|
410
545
|
"enabled": false,
|
|
411
|
-
"functionCall":
|
|
546
|
+
"functionCall": true,
|
|
412
547
|
"id": "meta-llama/llama-3.1-70b-instruct",
|
|
413
548
|
"maxTokens": undefined,
|
|
549
|
+
"pricing": {
|
|
550
|
+
"input": 0.3,
|
|
551
|
+
"output": 0.3,
|
|
552
|
+
},
|
|
414
553
|
"reasoning": false,
|
|
554
|
+
"releasedAt": "2024-07-23",
|
|
415
555
|
"vision": false,
|
|
416
556
|
},
|
|
417
557
|
{
|
|
@@ -428,7 +568,12 @@ _These are free, rate-limited endpoints for [Llama 3.1 8B Instruct](/models/meta
|
|
|
428
568
|
"functionCall": false,
|
|
429
569
|
"id": "meta-llama/llama-3.1-8b-instruct:free",
|
|
430
570
|
"maxTokens": 4096,
|
|
571
|
+
"pricing": {
|
|
572
|
+
"input": 0,
|
|
573
|
+
"output": 0,
|
|
574
|
+
},
|
|
431
575
|
"reasoning": false,
|
|
576
|
+
"releasedAt": "2024-07-23",
|
|
432
577
|
"vision": false,
|
|
433
578
|
},
|
|
434
579
|
{
|
|
@@ -440,10 +585,15 @@ It has demonstrated strong performance compared to leading closed-source models
|
|
|
440
585
|
To read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).",
|
|
441
586
|
"displayName": "Meta: Llama 3.1 8B Instruct",
|
|
442
587
|
"enabled": false,
|
|
443
|
-
"functionCall":
|
|
588
|
+
"functionCall": true,
|
|
444
589
|
"id": "meta-llama/llama-3.1-8b-instruct",
|
|
445
590
|
"maxTokens": undefined,
|
|
591
|
+
"pricing": {
|
|
592
|
+
"input": 0.055,
|
|
593
|
+
"output": 0.055,
|
|
594
|
+
},
|
|
446
595
|
"reasoning": false,
|
|
596
|
+
"releasedAt": "2024-07-23",
|
|
447
597
|
"vision": false,
|
|
448
598
|
},
|
|
449
599
|
{
|
|
@@ -457,10 +607,15 @@ It has demonstrated strong performance compared to leading closed-source models
|
|
|
457
607
|
To read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).",
|
|
458
608
|
"displayName": "Meta: Llama 3.1 405B Instruct",
|
|
459
609
|
"enabled": false,
|
|
460
|
-
"functionCall":
|
|
610
|
+
"functionCall": true,
|
|
461
611
|
"id": "meta-llama/llama-3.1-405b-instruct",
|
|
462
612
|
"maxTokens": undefined,
|
|
613
|
+
"pricing": {
|
|
614
|
+
"input": 1.79,
|
|
615
|
+
"output": 1.79,
|
|
616
|
+
},
|
|
463
617
|
"reasoning": false,
|
|
618
|
+
"releasedAt": "2024-07-23",
|
|
464
619
|
"vision": false,
|
|
465
620
|
},
|
|
466
621
|
{
|
|
@@ -475,7 +630,12 @@ Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.m
|
|
|
475
630
|
"functionCall": true,
|
|
476
631
|
"id": "cognitivecomputations/dolphin-llama-3-70b",
|
|
477
632
|
"maxTokens": undefined,
|
|
633
|
+
"pricing": {
|
|
634
|
+
"input": 0.35,
|
|
635
|
+
"output": 0.4,
|
|
636
|
+
},
|
|
478
637
|
"reasoning": false,
|
|
638
|
+
"releasedAt": "2024-07-19",
|
|
479
639
|
"vision": false,
|
|
480
640
|
},
|
|
481
641
|
{
|
|
@@ -489,10 +649,15 @@ Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.m
|
|
|
489
649
|
- Available under the Apache 2.0 license for free use, modification, and distribution",
|
|
490
650
|
"displayName": "Mistral: Codestral Mamba",
|
|
491
651
|
"enabled": false,
|
|
492
|
-
"functionCall":
|
|
652
|
+
"functionCall": true,
|
|
493
653
|
"id": "mistralai/codestral-mamba",
|
|
494
654
|
"maxTokens": undefined,
|
|
655
|
+
"pricing": {
|
|
656
|
+
"input": 0.25,
|
|
657
|
+
"output": 0.25,
|
|
658
|
+
},
|
|
495
659
|
"reasoning": false,
|
|
660
|
+
"releasedAt": "2024-07-19",
|
|
496
661
|
"vision": false,
|
|
497
662
|
},
|
|
498
663
|
{
|
|
@@ -507,7 +672,12 @@ It supports function calling and is released under the Apache 2.0 license.",
|
|
|
507
672
|
"functionCall": true,
|
|
508
673
|
"id": "mistralai/mistral-nemo",
|
|
509
674
|
"maxTokens": undefined,
|
|
675
|
+
"pricing": {
|
|
676
|
+
"input": 0.13,
|
|
677
|
+
"output": 0.13,
|
|
678
|
+
},
|
|
510
679
|
"reasoning": false,
|
|
680
|
+
"releasedAt": "2024-07-19",
|
|
511
681
|
"vision": false,
|
|
512
682
|
},
|
|
513
683
|
{
|
|
@@ -521,11 +691,16 @@ GPT-4o mini achieves an 82% score on MMLU and presently ranks higher than GPT-4
|
|
|
521
691
|
Check out the [launch announcement](https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/) to learn more.",
|
|
522
692
|
"displayName": "OpenAI: GPT-4o-mini (2024-07-18)",
|
|
523
693
|
"enabled": false,
|
|
524
|
-
"functionCall":
|
|
694
|
+
"functionCall": true,
|
|
525
695
|
"id": "openai/gpt-4o-mini-2024-07-18",
|
|
526
696
|
"maxTokens": 16384,
|
|
697
|
+
"pricing": {
|
|
698
|
+
"input": 0.15,
|
|
699
|
+
"output": 0.6,
|
|
700
|
+
},
|
|
527
701
|
"reasoning": false,
|
|
528
|
-
"
|
|
702
|
+
"releasedAt": "2024-07-18",
|
|
703
|
+
"vision": true,
|
|
529
704
|
},
|
|
530
705
|
{
|
|
531
706
|
"contextWindowTokens": 128000,
|
|
@@ -541,7 +716,12 @@ Check out the [launch announcement](https://openai.com/index/gpt-4o-mini-advanci
|
|
|
541
716
|
"functionCall": true,
|
|
542
717
|
"id": "openai/gpt-4o-mini",
|
|
543
718
|
"maxTokens": 16384,
|
|
719
|
+
"pricing": {
|
|
720
|
+
"input": 0.15,
|
|
721
|
+
"output": 0.6,
|
|
722
|
+
},
|
|
544
723
|
"reasoning": false,
|
|
724
|
+
"releasedAt": "2024-07-18",
|
|
545
725
|
"vision": true,
|
|
546
726
|
},
|
|
547
727
|
{
|
|
@@ -560,7 +740,12 @@ _These are free, rate-limited endpoints for [Qwen 2 7B Instruct](/models/qwen/qw
|
|
|
560
740
|
"functionCall": false,
|
|
561
741
|
"id": "qwen/qwen-2-7b-instruct:free",
|
|
562
742
|
"maxTokens": 4096,
|
|
743
|
+
"pricing": {
|
|
744
|
+
"input": 0,
|
|
745
|
+
"output": 0,
|
|
746
|
+
},
|
|
563
747
|
"reasoning": false,
|
|
748
|
+
"releasedAt": "2024-07-16",
|
|
564
749
|
"vision": false,
|
|
565
750
|
},
|
|
566
751
|
{
|
|
@@ -577,7 +762,12 @@ Usage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://hug
|
|
|
577
762
|
"functionCall": false,
|
|
578
763
|
"id": "qwen/qwen-2-7b-instruct",
|
|
579
764
|
"maxTokens": undefined,
|
|
765
|
+
"pricing": {
|
|
766
|
+
"input": 0.055,
|
|
767
|
+
"output": 0.055,
|
|
768
|
+
},
|
|
580
769
|
"reasoning": false,
|
|
770
|
+
"releasedAt": "2024-07-16",
|
|
581
771
|
"vision": false,
|
|
582
772
|
},
|
|
583
773
|
{
|
|
@@ -592,7 +782,12 @@ See the [launch announcement](https://blog.google/technology/developers/google-g
|
|
|
592
782
|
"functionCall": false,
|
|
593
783
|
"id": "google/gemma-2-27b-it",
|
|
594
784
|
"maxTokens": undefined,
|
|
785
|
+
"pricing": {
|
|
786
|
+
"input": 0.27,
|
|
787
|
+
"output": 0.27,
|
|
788
|
+
},
|
|
595
789
|
"reasoning": false,
|
|
790
|
+
"releasedAt": "2024-07-13",
|
|
596
791
|
"vision": false,
|
|
597
792
|
},
|
|
598
793
|
{
|
|
@@ -605,7 +800,12 @@ The model is based on [Qwen2 72B](https://openrouter.ai/models/qwen/qwen-2-72b-i
|
|
|
605
800
|
"functionCall": false,
|
|
606
801
|
"id": "alpindale/magnum-72b",
|
|
607
802
|
"maxTokens": 1024,
|
|
803
|
+
"pricing": {
|
|
804
|
+
"input": 3.75,
|
|
805
|
+
"output": 4.5,
|
|
806
|
+
},
|
|
608
807
|
"reasoning": false,
|
|
808
|
+
"releasedAt": "2024-07-11",
|
|
609
809
|
"vision": false,
|
|
610
810
|
},
|
|
611
811
|
{
|
|
@@ -618,7 +818,12 @@ Hermes-2 Θ (theta) was specifically designed with a few capabilities in mind: e
|
|
|
618
818
|
"functionCall": false,
|
|
619
819
|
"id": "nousresearch/hermes-2-theta-llama-3-8b",
|
|
620
820
|
"maxTokens": 2048,
|
|
821
|
+
"pricing": {
|
|
822
|
+
"input": 0.1875,
|
|
823
|
+
"output": 1.125,
|
|
824
|
+
},
|
|
621
825
|
"reasoning": false,
|
|
826
|
+
"releasedAt": "2024-07-11",
|
|
622
827
|
"vision": false,
|
|
623
828
|
},
|
|
624
829
|
{
|
|
@@ -635,7 +840,12 @@ _These are free, rate-limited endpoints for [Gemma 2 9B](/models/google/gemma-2-
|
|
|
635
840
|
"functionCall": false,
|
|
636
841
|
"id": "google/gemma-2-9b-it:free",
|
|
637
842
|
"maxTokens": 2048,
|
|
843
|
+
"pricing": {
|
|
844
|
+
"input": 0,
|
|
845
|
+
"output": 0,
|
|
846
|
+
},
|
|
638
847
|
"reasoning": false,
|
|
848
|
+
"releasedAt": "2024-06-28",
|
|
639
849
|
"vision": false,
|
|
640
850
|
},
|
|
641
851
|
{
|
|
@@ -650,7 +860,12 @@ See the [launch announcement](https://blog.google/technology/developers/google-g
|
|
|
650
860
|
"functionCall": false,
|
|
651
861
|
"id": "google/gemma-2-9b-it",
|
|
652
862
|
"maxTokens": undefined,
|
|
863
|
+
"pricing": {
|
|
864
|
+
"input": 0.06,
|
|
865
|
+
"output": 0.06,
|
|
866
|
+
},
|
|
653
867
|
"reasoning": false,
|
|
868
|
+
"releasedAt": "2024-06-28",
|
|
654
869
|
"vision": false,
|
|
655
870
|
},
|
|
656
871
|
{
|
|
@@ -666,7 +881,12 @@ Compared to older Stheno version, this model is trained on:
|
|
|
666
881
|
"functionCall": false,
|
|
667
882
|
"id": "sao10k/l3-stheno-8b",
|
|
668
883
|
"maxTokens": undefined,
|
|
884
|
+
"pricing": {
|
|
885
|
+
"input": 0.25,
|
|
886
|
+
"output": 1.5,
|
|
887
|
+
},
|
|
669
888
|
"reasoning": false,
|
|
889
|
+
"releasedAt": "2024-06-27",
|
|
670
890
|
"vision": false,
|
|
671
891
|
},
|
|
672
892
|
{
|
|
@@ -684,7 +904,12 @@ Jamba has a knowledge cutoff of February 2024.",
|
|
|
684
904
|
"functionCall": false,
|
|
685
905
|
"id": "ai21/jamba-instruct",
|
|
686
906
|
"maxTokens": 4096,
|
|
907
|
+
"pricing": {
|
|
908
|
+
"input": 0.5,
|
|
909
|
+
"output": 0.7,
|
|
910
|
+
},
|
|
687
911
|
"reasoning": false,
|
|
912
|
+
"releasedAt": "2024-06-25",
|
|
688
913
|
"vision": false,
|
|
689
914
|
},
|
|
690
915
|
{
|
|
@@ -702,7 +927,12 @@ Jamba has a knowledge cutoff of February 2024.",
|
|
|
702
927
|
"functionCall": true,
|
|
703
928
|
"id": "anthropic/claude-3.5-sonnet",
|
|
704
929
|
"maxTokens": 8192,
|
|
930
|
+
"pricing": {
|
|
931
|
+
"input": 3,
|
|
932
|
+
"output": 15,
|
|
933
|
+
},
|
|
705
934
|
"reasoning": false,
|
|
935
|
+
"releasedAt": "2024-06-20",
|
|
706
936
|
"vision": true,
|
|
707
937
|
},
|
|
708
938
|
{
|
|
@@ -722,7 +952,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
722
952
|
"functionCall": true,
|
|
723
953
|
"id": "anthropic/claude-3.5-sonnet:beta",
|
|
724
954
|
"maxTokens": 8192,
|
|
955
|
+
"pricing": {
|
|
956
|
+
"input": 3,
|
|
957
|
+
"output": 15,
|
|
958
|
+
},
|
|
725
959
|
"reasoning": false,
|
|
960
|
+
"releasedAt": "2024-06-20",
|
|
726
961
|
"vision": true,
|
|
727
962
|
},
|
|
728
963
|
{
|
|
@@ -739,7 +974,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
739
974
|
"functionCall": false,
|
|
740
975
|
"id": "sao10k/l3-euryale-70b",
|
|
741
976
|
"maxTokens": undefined,
|
|
977
|
+
"pricing": {
|
|
978
|
+
"input": 0.35,
|
|
979
|
+
"output": 0.4,
|
|
980
|
+
},
|
|
742
981
|
"reasoning": false,
|
|
982
|
+
"releasedAt": "2024-06-18",
|
|
743
983
|
"vision": false,
|
|
744
984
|
},
|
|
745
985
|
{
|
|
@@ -754,7 +994,12 @@ For 128k context length, try [Phi-3 Medium 128K](/models/microsoft/phi-3-medium-
|
|
|
754
994
|
"functionCall": false,
|
|
755
995
|
"id": "microsoft/phi-3-medium-4k-instruct",
|
|
756
996
|
"maxTokens": undefined,
|
|
997
|
+
"pricing": {
|
|
998
|
+
"input": 0.14,
|
|
999
|
+
"output": 0.14,
|
|
1000
|
+
},
|
|
757
1001
|
"reasoning": false,
|
|
1002
|
+
"releasedAt": "2024-06-15",
|
|
758
1003
|
"vision": false,
|
|
759
1004
|
},
|
|
760
1005
|
{
|
|
@@ -771,7 +1016,12 @@ The model is uncensored and is stripped of alignment and bias. It requires an ex
|
|
|
771
1016
|
"functionCall": false,
|
|
772
1017
|
"id": "cognitivecomputations/dolphin-mixtral-8x22b",
|
|
773
1018
|
"maxTokens": undefined,
|
|
1019
|
+
"pricing": {
|
|
1020
|
+
"input": 0.9,
|
|
1021
|
+
"output": 0.9,
|
|
1022
|
+
},
|
|
774
1023
|
"reasoning": false,
|
|
1024
|
+
"releasedAt": "2024-06-08",
|
|
775
1025
|
"vision": false,
|
|
776
1026
|
},
|
|
777
1027
|
{
|
|
@@ -788,7 +1038,12 @@ Usage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://hug
|
|
|
788
1038
|
"functionCall": false,
|
|
789
1039
|
"id": "qwen/qwen-2-72b-instruct",
|
|
790
1040
|
"maxTokens": undefined,
|
|
1041
|
+
"pricing": {
|
|
1042
|
+
"input": 0.35,
|
|
1043
|
+
"output": 0.4,
|
|
1044
|
+
},
|
|
791
1045
|
"reasoning": false,
|
|
1046
|
+
"releasedAt": "2024-06-07",
|
|
792
1047
|
"vision": false,
|
|
793
1048
|
},
|
|
794
1049
|
{
|
|
@@ -806,7 +1061,12 @@ It outperforms many similarly sized models including [Llama 3 8B Instruct](/mode
|
|
|
806
1061
|
"functionCall": false,
|
|
807
1062
|
"id": "openchat/openchat-8b",
|
|
808
1063
|
"maxTokens": undefined,
|
|
1064
|
+
"pricing": {
|
|
1065
|
+
"input": 0.055,
|
|
1066
|
+
"output": 0.055,
|
|
1067
|
+
},
|
|
809
1068
|
"reasoning": false,
|
|
1069
|
+
"releasedAt": "2024-06-01",
|
|
810
1070
|
"vision": false,
|
|
811
1071
|
},
|
|
812
1072
|
{
|
|
@@ -817,7 +1077,12 @@ It outperforms many similarly sized models including [Llama 3 8B Instruct](/mode
|
|
|
817
1077
|
"functionCall": false,
|
|
818
1078
|
"id": "nousresearch/hermes-2-pro-llama-3-8b",
|
|
819
1079
|
"maxTokens": undefined,
|
|
1080
|
+
"pricing": {
|
|
1081
|
+
"input": 0.14,
|
|
1082
|
+
"output": 0.14,
|
|
1083
|
+
},
|
|
820
1084
|
"reasoning": false,
|
|
1085
|
+
"releasedAt": "2024-05-27",
|
|
821
1086
|
"vision": false,
|
|
822
1087
|
},
|
|
823
1088
|
{
|
|
@@ -836,7 +1101,12 @@ NOTE: Support for function calling depends on the provider.",
|
|
|
836
1101
|
"functionCall": true,
|
|
837
1102
|
"id": "mistralai/mistral-7b-instruct-v0.3",
|
|
838
1103
|
"maxTokens": undefined,
|
|
1104
|
+
"pricing": {
|
|
1105
|
+
"input": 0.055,
|
|
1106
|
+
"output": 0.055,
|
|
1107
|
+
},
|
|
839
1108
|
"reasoning": false,
|
|
1109
|
+
"releasedAt": "2024-05-27",
|
|
840
1110
|
"vision": false,
|
|
841
1111
|
},
|
|
842
1112
|
{
|
|
@@ -851,7 +1121,12 @@ _These are free, rate-limited endpoints for [Mistral 7B Instruct](/models/mistra
|
|
|
851
1121
|
"functionCall": false,
|
|
852
1122
|
"id": "mistralai/mistral-7b-instruct:free",
|
|
853
1123
|
"maxTokens": 4096,
|
|
1124
|
+
"pricing": {
|
|
1125
|
+
"input": 0,
|
|
1126
|
+
"output": 0,
|
|
1127
|
+
},
|
|
854
1128
|
"reasoning": false,
|
|
1129
|
+
"releasedAt": "2024-05-27",
|
|
855
1130
|
"vision": false,
|
|
856
1131
|
},
|
|
857
1132
|
{
|
|
@@ -861,10 +1136,15 @@ _These are free, rate-limited endpoints for [Mistral 7B Instruct](/models/mistra
|
|
|
861
1136
|
*Mistral 7B Instruct has multiple version variants, and this is intended to be the latest version.*",
|
|
862
1137
|
"displayName": "Mistral: Mistral 7B Instruct",
|
|
863
1138
|
"enabled": true,
|
|
864
|
-
"functionCall":
|
|
1139
|
+
"functionCall": true,
|
|
865
1140
|
"id": "mistralai/mistral-7b-instruct",
|
|
866
1141
|
"maxTokens": undefined,
|
|
1142
|
+
"pricing": {
|
|
1143
|
+
"input": 0.055,
|
|
1144
|
+
"output": 0.055,
|
|
1145
|
+
},
|
|
867
1146
|
"reasoning": false,
|
|
1147
|
+
"releasedAt": "2024-05-27",
|
|
868
1148
|
"vision": false,
|
|
869
1149
|
},
|
|
870
1150
|
{
|
|
@@ -879,7 +1159,12 @@ _These are higher-throughput endpoints for [Mistral 7B Instruct](/models/mistral
|
|
|
879
1159
|
"functionCall": false,
|
|
880
1160
|
"id": "mistralai/mistral-7b-instruct:nitro",
|
|
881
1161
|
"maxTokens": undefined,
|
|
1162
|
+
"pricing": {
|
|
1163
|
+
"input": 0.07,
|
|
1164
|
+
"output": 0.07,
|
|
1165
|
+
},
|
|
882
1166
|
"reasoning": false,
|
|
1167
|
+
"releasedAt": "2024-05-27",
|
|
883
1168
|
"vision": false,
|
|
884
1169
|
},
|
|
885
1170
|
{
|
|
@@ -891,10 +1176,15 @@ At time of release, Phi-3 Medium demonstrated state-of-the-art performance among
|
|
|
891
1176
|
_These are free, rate-limited endpoints for [Phi-3 Mini 128K Instruct](/models/microsoft/phi-3-mini-128k-instruct). Outputs may be cached. Read about rate limits [here](/docs/limits)._",
|
|
892
1177
|
"displayName": "Phi-3 Mini 128K Instruct (free)",
|
|
893
1178
|
"enabled": false,
|
|
894
|
-
"functionCall":
|
|
1179
|
+
"functionCall": true,
|
|
895
1180
|
"id": "microsoft/phi-3-mini-128k-instruct:free",
|
|
896
1181
|
"maxTokens": 4096,
|
|
1182
|
+
"pricing": {
|
|
1183
|
+
"input": 0,
|
|
1184
|
+
"output": 0,
|
|
1185
|
+
},
|
|
897
1186
|
"reasoning": false,
|
|
1187
|
+
"releasedAt": "2024-05-26",
|
|
898
1188
|
"vision": false,
|
|
899
1189
|
},
|
|
900
1190
|
{
|
|
@@ -904,10 +1194,15 @@ _These are free, rate-limited endpoints for [Phi-3 Mini 128K Instruct](/models/m
|
|
|
904
1194
|
At time of release, Phi-3 Medium demonstrated state-of-the-art performance among lightweight models. This model is static, trained on an offline dataset with an October 2023 cutoff date.",
|
|
905
1195
|
"displayName": "Phi-3 Mini 128K Instruct",
|
|
906
1196
|
"enabled": false,
|
|
907
|
-
"functionCall":
|
|
1197
|
+
"functionCall": true,
|
|
908
1198
|
"id": "microsoft/phi-3-mini-128k-instruct",
|
|
909
1199
|
"maxTokens": undefined,
|
|
1200
|
+
"pricing": {
|
|
1201
|
+
"input": 0.1,
|
|
1202
|
+
"output": 0.1,
|
|
1203
|
+
},
|
|
910
1204
|
"reasoning": false,
|
|
1205
|
+
"releasedAt": "2024-05-26",
|
|
911
1206
|
"vision": false,
|
|
912
1207
|
},
|
|
913
1208
|
{
|
|
@@ -921,10 +1216,15 @@ For 4k context length, try [Phi-3 Medium 4K](/models/microsoft/phi-3-medium-4k-i
|
|
|
921
1216
|
_These are free, rate-limited endpoints for [Phi-3 Medium 128K Instruct](/models/microsoft/phi-3-medium-128k-instruct). Outputs may be cached. Read about rate limits [here](/docs/limits)._",
|
|
922
1217
|
"displayName": "Phi-3 Medium 128K Instruct (free)",
|
|
923
1218
|
"enabled": false,
|
|
924
|
-
"functionCall":
|
|
1219
|
+
"functionCall": true,
|
|
925
1220
|
"id": "microsoft/phi-3-medium-128k-instruct:free",
|
|
926
1221
|
"maxTokens": 4096,
|
|
1222
|
+
"pricing": {
|
|
1223
|
+
"input": 0,
|
|
1224
|
+
"output": 0,
|
|
1225
|
+
},
|
|
927
1226
|
"reasoning": false,
|
|
1227
|
+
"releasedAt": "2024-05-24",
|
|
928
1228
|
"vision": false,
|
|
929
1229
|
},
|
|
930
1230
|
{
|
|
@@ -936,10 +1236,15 @@ At time of release, Phi-3 Medium demonstrated state-of-the-art performance among
|
|
|
936
1236
|
For 4k context length, try [Phi-3 Medium 4K](/models/microsoft/phi-3-medium-4k-instruct).",
|
|
937
1237
|
"displayName": "Phi-3 Medium 128K Instruct",
|
|
938
1238
|
"enabled": false,
|
|
939
|
-
"functionCall":
|
|
1239
|
+
"functionCall": true,
|
|
940
1240
|
"id": "microsoft/phi-3-medium-128k-instruct",
|
|
941
1241
|
"maxTokens": undefined,
|
|
1242
|
+
"pricing": {
|
|
1243
|
+
"input": 1,
|
|
1244
|
+
"output": 1,
|
|
1245
|
+
},
|
|
942
1246
|
"reasoning": false,
|
|
1247
|
+
"releasedAt": "2024-05-24",
|
|
943
1248
|
"vision": false,
|
|
944
1249
|
},
|
|
945
1250
|
{
|
|
@@ -954,7 +1259,12 @@ Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.m
|
|
|
954
1259
|
"functionCall": false,
|
|
955
1260
|
"id": "neversleep/llama-3-lumimaid-70b",
|
|
956
1261
|
"maxTokens": 2048,
|
|
1262
|
+
"pricing": {
|
|
1263
|
+
"input": 3.375,
|
|
1264
|
+
"output": 4.5,
|
|
1265
|
+
},
|
|
957
1266
|
"reasoning": false,
|
|
1267
|
+
"releasedAt": "2024-05-16",
|
|
958
1268
|
"vision": false,
|
|
959
1269
|
},
|
|
960
1270
|
{
|
|
@@ -971,7 +1281,12 @@ Usage of Gemini is subject to Google's [Gemini Terms of Use](https://ai.google.d
|
|
|
971
1281
|
"functionCall": true,
|
|
972
1282
|
"id": "google/gemini-flash-1.5",
|
|
973
1283
|
"maxTokens": 32768,
|
|
1284
|
+
"pricing": {
|
|
1285
|
+
"input": 0.0375,
|
|
1286
|
+
"output": 0.15,
|
|
1287
|
+
},
|
|
974
1288
|
"reasoning": false,
|
|
1289
|
+
"releasedAt": "2024-05-14",
|
|
975
1290
|
"vision": true,
|
|
976
1291
|
},
|
|
977
1292
|
{
|
|
@@ -984,7 +1299,12 @@ The original V1 model was trained from scratch on 2T tokens, with a composition
|
|
|
984
1299
|
"functionCall": false,
|
|
985
1300
|
"id": "deepseek/deepseek-coder",
|
|
986
1301
|
"maxTokens": 4096,
|
|
1302
|
+
"pricing": {
|
|
1303
|
+
"input": 0.14,
|
|
1304
|
+
"output": 0.28,
|
|
1305
|
+
},
|
|
987
1306
|
"reasoning": false,
|
|
1307
|
+
"releasedAt": "2024-05-14",
|
|
988
1308
|
"vision": false,
|
|
989
1309
|
},
|
|
990
1310
|
{
|
|
@@ -999,7 +1319,12 @@ DeepSeek-V2 achieves remarkable performance on both standard benchmarks and open
|
|
|
999
1319
|
"functionCall": true,
|
|
1000
1320
|
"id": "deepseek/deepseek-chat",
|
|
1001
1321
|
"maxTokens": 4096,
|
|
1322
|
+
"pricing": {
|
|
1323
|
+
"input": 0.14,
|
|
1324
|
+
"output": 0.28,
|
|
1325
|
+
},
|
|
1002
1326
|
"reasoning": false,
|
|
1327
|
+
"releasedAt": "2024-05-14",
|
|
1003
1328
|
"vision": false,
|
|
1004
1329
|
},
|
|
1005
1330
|
{
|
|
@@ -1012,7 +1337,12 @@ This is the online version of the [offline chat model](/models/perplexity/llama-
|
|
|
1012
1337
|
"functionCall": false,
|
|
1013
1338
|
"id": "perplexity/llama-3-sonar-large-32k-online",
|
|
1014
1339
|
"maxTokens": undefined,
|
|
1340
|
+
"pricing": {
|
|
1341
|
+
"input": 1,
|
|
1342
|
+
"output": 1,
|
|
1343
|
+
},
|
|
1015
1344
|
"reasoning": false,
|
|
1345
|
+
"releasedAt": "2024-05-14",
|
|
1016
1346
|
"vision": false,
|
|
1017
1347
|
},
|
|
1018
1348
|
{
|
|
@@ -1025,7 +1355,12 @@ This is a normal offline LLM, but the [online version](/models/perplexity/llama-
|
|
|
1025
1355
|
"functionCall": false,
|
|
1026
1356
|
"id": "perplexity/llama-3-sonar-large-32k-chat",
|
|
1027
1357
|
"maxTokens": undefined,
|
|
1358
|
+
"pricing": {
|
|
1359
|
+
"input": 1,
|
|
1360
|
+
"output": 1,
|
|
1361
|
+
},
|
|
1028
1362
|
"reasoning": false,
|
|
1363
|
+
"releasedAt": "2024-05-14",
|
|
1029
1364
|
"vision": false,
|
|
1030
1365
|
},
|
|
1031
1366
|
{
|
|
@@ -1038,7 +1373,12 @@ This is the online version of the [offline chat model](/models/perplexity/llama-
|
|
|
1038
1373
|
"functionCall": false,
|
|
1039
1374
|
"id": "perplexity/llama-3-sonar-small-32k-online",
|
|
1040
1375
|
"maxTokens": undefined,
|
|
1376
|
+
"pricing": {
|
|
1377
|
+
"input": 0.2,
|
|
1378
|
+
"output": 0.2,
|
|
1379
|
+
},
|
|
1041
1380
|
"reasoning": false,
|
|
1381
|
+
"releasedAt": "2024-05-14",
|
|
1042
1382
|
"vision": false,
|
|
1043
1383
|
},
|
|
1044
1384
|
{
|
|
@@ -1051,7 +1391,12 @@ This is a normal offline LLM, but the [online version](/models/perplexity/llama-
|
|
|
1051
1391
|
"functionCall": false,
|
|
1052
1392
|
"id": "perplexity/llama-3-sonar-small-32k-chat",
|
|
1053
1393
|
"maxTokens": undefined,
|
|
1394
|
+
"pricing": {
|
|
1395
|
+
"input": 0.2,
|
|
1396
|
+
"output": 0.2,
|
|
1397
|
+
},
|
|
1054
1398
|
"reasoning": false,
|
|
1399
|
+
"releasedAt": "2024-05-14",
|
|
1055
1400
|
"vision": false,
|
|
1056
1401
|
},
|
|
1057
1402
|
{
|
|
@@ -1070,7 +1415,12 @@ To read more about the model release, [click here](https://ai.meta.com/blog/meta
|
|
|
1070
1415
|
"functionCall": false,
|
|
1071
1416
|
"id": "meta-llama/llama-guard-2-8b",
|
|
1072
1417
|
"maxTokens": undefined,
|
|
1418
|
+
"pricing": {
|
|
1419
|
+
"input": 0.18,
|
|
1420
|
+
"output": 0.18,
|
|
1421
|
+
},
|
|
1073
1422
|
"reasoning": false,
|
|
1423
|
+
"releasedAt": "2024-05-13",
|
|
1074
1424
|
"vision": false,
|
|
1075
1425
|
},
|
|
1076
1426
|
{
|
|
@@ -1080,11 +1430,16 @@ To read more about the model release, [click here](https://ai.meta.com/blog/meta
|
|
|
1080
1430
|
For benchmarking against other models, it was briefly called ["im-also-a-good-gpt2-chatbot"](https://twitter.com/LiamFedus/status/1790064963966370209)",
|
|
1081
1431
|
"displayName": "OpenAI: GPT-4o (2024-05-13)",
|
|
1082
1432
|
"enabled": false,
|
|
1083
|
-
"functionCall":
|
|
1433
|
+
"functionCall": true,
|
|
1084
1434
|
"id": "openai/gpt-4o-2024-05-13",
|
|
1085
1435
|
"maxTokens": 4096,
|
|
1436
|
+
"pricing": {
|
|
1437
|
+
"input": 5,
|
|
1438
|
+
"output": 15,
|
|
1439
|
+
},
|
|
1086
1440
|
"reasoning": false,
|
|
1087
|
-
"
|
|
1441
|
+
"releasedAt": "2024-05-13",
|
|
1442
|
+
"vision": true,
|
|
1088
1443
|
},
|
|
1089
1444
|
{
|
|
1090
1445
|
"contextWindowTokens": 128000,
|
|
@@ -1096,7 +1451,12 @@ For benchmarking against other models, it was briefly called ["im-also-a-good-gp
|
|
|
1096
1451
|
"functionCall": true,
|
|
1097
1452
|
"id": "openai/gpt-4o",
|
|
1098
1453
|
"maxTokens": 4096,
|
|
1454
|
+
"pricing": {
|
|
1455
|
+
"input": 5,
|
|
1456
|
+
"output": 15,
|
|
1457
|
+
},
|
|
1099
1458
|
"reasoning": false,
|
|
1459
|
+
"releasedAt": "2024-05-13",
|
|
1100
1460
|
"vision": true,
|
|
1101
1461
|
},
|
|
1102
1462
|
{
|
|
@@ -1106,10 +1466,15 @@ For benchmarking against other models, it was briefly called ["im-also-a-good-gp
|
|
|
1106
1466
|
_These are extended-context endpoints for [GPT-4o](/models/openai/gpt-4o). They may have higher prices._",
|
|
1107
1467
|
"displayName": "OpenAI: GPT-4o (extended)",
|
|
1108
1468
|
"enabled": false,
|
|
1109
|
-
"functionCall":
|
|
1469
|
+
"functionCall": true,
|
|
1110
1470
|
"id": "openai/gpt-4o:extended",
|
|
1111
1471
|
"maxTokens": 64000,
|
|
1472
|
+
"pricing": {
|
|
1473
|
+
"input": 6,
|
|
1474
|
+
"output": 18,
|
|
1475
|
+
},
|
|
1112
1476
|
"reasoning": false,
|
|
1477
|
+
"releasedAt": "2024-05-13",
|
|
1113
1478
|
"vision": false,
|
|
1114
1479
|
},
|
|
1115
1480
|
{
|
|
@@ -1128,7 +1493,12 @@ Usage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://hug
|
|
|
1128
1493
|
"functionCall": false,
|
|
1129
1494
|
"id": "qwen/qwen-72b-chat",
|
|
1130
1495
|
"maxTokens": undefined,
|
|
1496
|
+
"pricing": {
|
|
1497
|
+
"input": 0.81,
|
|
1498
|
+
"output": 0.81,
|
|
1499
|
+
},
|
|
1131
1500
|
"reasoning": false,
|
|
1501
|
+
"releasedAt": "2024-05-09",
|
|
1132
1502
|
"vision": false,
|
|
1133
1503
|
},
|
|
1134
1504
|
{
|
|
@@ -1147,7 +1517,12 @@ Usage of this model is subject to [Tongyi Qianwen LICENSE AGREEMENT](https://hug
|
|
|
1147
1517
|
"functionCall": false,
|
|
1148
1518
|
"id": "qwen/qwen-110b-chat",
|
|
1149
1519
|
"maxTokens": undefined,
|
|
1520
|
+
"pricing": {
|
|
1521
|
+
"input": 1.62,
|
|
1522
|
+
"output": 1.62,
|
|
1523
|
+
},
|
|
1150
1524
|
"reasoning": false,
|
|
1525
|
+
"releasedAt": "2024-05-09",
|
|
1151
1526
|
"vision": false,
|
|
1152
1527
|
},
|
|
1153
1528
|
{
|
|
@@ -1162,7 +1537,12 @@ Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.m
|
|
|
1162
1537
|
"functionCall": false,
|
|
1163
1538
|
"id": "neversleep/llama-3-lumimaid-8b",
|
|
1164
1539
|
"maxTokens": undefined,
|
|
1540
|
+
"pricing": {
|
|
1541
|
+
"input": 0.1875,
|
|
1542
|
+
"output": 1.125,
|
|
1543
|
+
},
|
|
1165
1544
|
"reasoning": false,
|
|
1545
|
+
"releasedAt": "2024-05-04",
|
|
1166
1546
|
"vision": false,
|
|
1167
1547
|
},
|
|
1168
1548
|
{
|
|
@@ -1179,7 +1559,12 @@ _These are extended-context endpoints for [Llama 3 Lumimaid 8B](/models/neversle
|
|
|
1179
1559
|
"functionCall": false,
|
|
1180
1560
|
"id": "neversleep/llama-3-lumimaid-8b:extended",
|
|
1181
1561
|
"maxTokens": 2048,
|
|
1562
|
+
"pricing": {
|
|
1563
|
+
"input": 0.1875,
|
|
1564
|
+
"output": 1.125,
|
|
1565
|
+
},
|
|
1182
1566
|
"reasoning": false,
|
|
1567
|
+
"releasedAt": "2024-05-04",
|
|
1183
1568
|
"vision": false,
|
|
1184
1569
|
},
|
|
1185
1570
|
{
|
|
@@ -1192,7 +1577,12 @@ If you submit a raw prompt, you can use Alpaca or Vicuna formats.",
|
|
|
1192
1577
|
"functionCall": false,
|
|
1193
1578
|
"id": "sao10k/fimbulvetr-11b-v2",
|
|
1194
1579
|
"maxTokens": 2048,
|
|
1580
|
+
"pricing": {
|
|
1581
|
+
"input": 0.375,
|
|
1582
|
+
"output": 1.5,
|
|
1583
|
+
},
|
|
1195
1584
|
"reasoning": false,
|
|
1585
|
+
"releasedAt": "2024-04-21",
|
|
1196
1586
|
"vision": false,
|
|
1197
1587
|
},
|
|
1198
1588
|
{
|
|
@@ -1204,10 +1594,15 @@ It has demonstrated strong performance compared to leading closed-source models
|
|
|
1204
1594
|
To read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).",
|
|
1205
1595
|
"displayName": "Meta: Llama 3 70B Instruct",
|
|
1206
1596
|
"enabled": false,
|
|
1207
|
-
"functionCall":
|
|
1597
|
+
"functionCall": true,
|
|
1208
1598
|
"id": "meta-llama/llama-3-70b-instruct",
|
|
1209
1599
|
"maxTokens": undefined,
|
|
1600
|
+
"pricing": {
|
|
1601
|
+
"input": 0.35,
|
|
1602
|
+
"output": 0.4,
|
|
1603
|
+
},
|
|
1210
1604
|
"reasoning": false,
|
|
1605
|
+
"releasedAt": "2024-04-18",
|
|
1211
1606
|
"vision": false,
|
|
1212
1607
|
},
|
|
1213
1608
|
{
|
|
@@ -1224,7 +1619,12 @@ _These are higher-throughput endpoints for [Llama 3 70B Instruct](/models/meta-l
|
|
|
1224
1619
|
"functionCall": false,
|
|
1225
1620
|
"id": "meta-llama/llama-3-70b-instruct:nitro",
|
|
1226
1621
|
"maxTokens": undefined,
|
|
1622
|
+
"pricing": {
|
|
1623
|
+
"input": 0.792,
|
|
1624
|
+
"output": 0.792,
|
|
1625
|
+
},
|
|
1227
1626
|
"reasoning": false,
|
|
1627
|
+
"releasedAt": "2024-04-18",
|
|
1228
1628
|
"vision": false,
|
|
1229
1629
|
},
|
|
1230
1630
|
{
|
|
@@ -1241,7 +1641,12 @@ _These are free, rate-limited endpoints for [Llama 3 8B Instruct](/models/meta-l
|
|
|
1241
1641
|
"functionCall": false,
|
|
1242
1642
|
"id": "meta-llama/llama-3-8b-instruct:free",
|
|
1243
1643
|
"maxTokens": 4096,
|
|
1644
|
+
"pricing": {
|
|
1645
|
+
"input": 0,
|
|
1646
|
+
"output": 0,
|
|
1647
|
+
},
|
|
1244
1648
|
"reasoning": false,
|
|
1649
|
+
"releasedAt": "2024-04-18",
|
|
1245
1650
|
"vision": false,
|
|
1246
1651
|
},
|
|
1247
1652
|
{
|
|
@@ -1253,10 +1658,15 @@ It has demonstrated strong performance compared to leading closed-source models
|
|
|
1253
1658
|
To read more about the model release, [click here](https://ai.meta.com/blog/meta-llama-3/). Usage of this model is subject to [Meta's Acceptable Use Policy](https://llama.meta.com/llama3/use-policy/).",
|
|
1254
1659
|
"displayName": "Meta: Llama 3 8B Instruct",
|
|
1255
1660
|
"enabled": false,
|
|
1256
|
-
"functionCall":
|
|
1661
|
+
"functionCall": true,
|
|
1257
1662
|
"id": "meta-llama/llama-3-8b-instruct",
|
|
1258
1663
|
"maxTokens": undefined,
|
|
1664
|
+
"pricing": {
|
|
1665
|
+
"input": 0.055,
|
|
1666
|
+
"output": 0.055,
|
|
1667
|
+
},
|
|
1259
1668
|
"reasoning": false,
|
|
1669
|
+
"releasedAt": "2024-04-18",
|
|
1260
1670
|
"vision": false,
|
|
1261
1671
|
},
|
|
1262
1672
|
{
|
|
@@ -1273,7 +1683,12 @@ _These are higher-throughput endpoints for [Llama 3 8B Instruct](/models/meta-ll
|
|
|
1273
1683
|
"functionCall": false,
|
|
1274
1684
|
"id": "meta-llama/llama-3-8b-instruct:nitro",
|
|
1275
1685
|
"maxTokens": undefined,
|
|
1686
|
+
"pricing": {
|
|
1687
|
+
"input": 0.162,
|
|
1688
|
+
"output": 0.162,
|
|
1689
|
+
},
|
|
1276
1690
|
"reasoning": false,
|
|
1691
|
+
"releasedAt": "2024-04-18",
|
|
1277
1692
|
"vision": false,
|
|
1278
1693
|
},
|
|
1279
1694
|
{
|
|
@@ -1290,7 +1705,12 @@ _These are extended-context endpoints for [Llama 3 8B Instruct](/models/meta-lla
|
|
|
1290
1705
|
"functionCall": false,
|
|
1291
1706
|
"id": "meta-llama/llama-3-8b-instruct:extended",
|
|
1292
1707
|
"maxTokens": 2048,
|
|
1708
|
+
"pricing": {
|
|
1709
|
+
"input": 0.1875,
|
|
1710
|
+
"output": 1.125,
|
|
1711
|
+
},
|
|
1293
1712
|
"reasoning": false,
|
|
1713
|
+
"releasedAt": "2024-04-18",
|
|
1294
1714
|
"vision": false,
|
|
1295
1715
|
},
|
|
1296
1716
|
{
|
|
@@ -1307,7 +1727,12 @@ See benchmarks on the launch announcement [here](https://mistral.ai/news/mixtral
|
|
|
1307
1727
|
"functionCall": false,
|
|
1308
1728
|
"id": "mistralai/mixtral-8x22b-instruct",
|
|
1309
1729
|
"maxTokens": undefined,
|
|
1730
|
+
"pricing": {
|
|
1731
|
+
"input": 0.65,
|
|
1732
|
+
"output": 0.65,
|
|
1733
|
+
},
|
|
1310
1734
|
"reasoning": false,
|
|
1735
|
+
"releasedAt": "2024-04-17",
|
|
1311
1736
|
"vision": false,
|
|
1312
1737
|
},
|
|
1313
1738
|
{
|
|
@@ -1324,7 +1749,12 @@ To read more about the model release, [click here](https://wizardlm.github.io/Wi
|
|
|
1324
1749
|
"functionCall": false,
|
|
1325
1750
|
"id": "microsoft/wizardlm-2-7b",
|
|
1326
1751
|
"maxTokens": undefined,
|
|
1752
|
+
"pricing": {
|
|
1753
|
+
"input": 0.055,
|
|
1754
|
+
"output": 0.055,
|
|
1755
|
+
},
|
|
1327
1756
|
"reasoning": false,
|
|
1757
|
+
"releasedAt": "2024-04-16",
|
|
1328
1758
|
"vision": false,
|
|
1329
1759
|
},
|
|
1330
1760
|
{
|
|
@@ -1341,7 +1771,12 @@ To read more about the model release, [click here](https://wizardlm.github.io/Wi
|
|
|
1341
1771
|
"functionCall": false,
|
|
1342
1772
|
"id": "microsoft/wizardlm-2-8x22b",
|
|
1343
1773
|
"maxTokens": undefined,
|
|
1774
|
+
"pricing": {
|
|
1775
|
+
"input": 0.5,
|
|
1776
|
+
"output": 0.5,
|
|
1777
|
+
},
|
|
1344
1778
|
"reasoning": false,
|
|
1779
|
+
"releasedAt": "2024-04-16",
|
|
1345
1780
|
"vision": false,
|
|
1346
1781
|
},
|
|
1347
1782
|
{
|
|
@@ -1367,7 +1802,12 @@ Usage of Gemini is subject to Google's [Gemini Terms of Use](https://ai.google.d
|
|
|
1367
1802
|
"functionCall": true,
|
|
1368
1803
|
"id": "google/gemini-pro-1.5",
|
|
1369
1804
|
"maxTokens": 32768,
|
|
1805
|
+
"pricing": {
|
|
1806
|
+
"input": 2.5,
|
|
1807
|
+
"output": 7.5,
|
|
1808
|
+
},
|
|
1370
1809
|
"reasoning": false,
|
|
1810
|
+
"releasedAt": "2024-04-09",
|
|
1371
1811
|
"vision": true,
|
|
1372
1812
|
},
|
|
1373
1813
|
{
|
|
@@ -1380,7 +1820,12 @@ Training data: up to December 2023.",
|
|
|
1380
1820
|
"functionCall": true,
|
|
1381
1821
|
"id": "openai/gpt-4-turbo",
|
|
1382
1822
|
"maxTokens": 4096,
|
|
1823
|
+
"pricing": {
|
|
1824
|
+
"input": 10,
|
|
1825
|
+
"output": 30,
|
|
1826
|
+
},
|
|
1383
1827
|
"reasoning": false,
|
|
1828
|
+
"releasedAt": "2024-04-09",
|
|
1384
1829
|
"vision": true,
|
|
1385
1830
|
},
|
|
1386
1831
|
{
|
|
@@ -1392,10 +1837,15 @@ It offers multilingual support for ten key languages to facilitate global busine
|
|
|
1392
1837
|
Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).",
|
|
1393
1838
|
"displayName": "Cohere: Command R+",
|
|
1394
1839
|
"enabled": false,
|
|
1395
|
-
"functionCall":
|
|
1840
|
+
"functionCall": true,
|
|
1396
1841
|
"id": "cohere/command-r-plus",
|
|
1397
1842
|
"maxTokens": 4000,
|
|
1843
|
+
"pricing": {
|
|
1844
|
+
"input": 3,
|
|
1845
|
+
"output": 15,
|
|
1846
|
+
},
|
|
1398
1847
|
"reasoning": false,
|
|
1848
|
+
"releasedAt": "2024-04-04",
|
|
1399
1849
|
"vision": false,
|
|
1400
1850
|
},
|
|
1401
1851
|
{
|
|
@@ -1412,7 +1862,12 @@ See the launch announcement and benchmark results [here](https://www.databricks.
|
|
|
1412
1862
|
"functionCall": false,
|
|
1413
1863
|
"id": "databricks/dbrx-instruct",
|
|
1414
1864
|
"maxTokens": undefined,
|
|
1865
|
+
"pricing": {
|
|
1866
|
+
"input": 1.08,
|
|
1867
|
+
"output": 1.08,
|
|
1868
|
+
},
|
|
1415
1869
|
"reasoning": false,
|
|
1870
|
+
"releasedAt": "2024-03-29",
|
|
1416
1871
|
"vision": false,
|
|
1417
1872
|
},
|
|
1418
1873
|
{
|
|
@@ -1425,7 +1880,12 @@ Descending from earlier versions of Midnight Rose and [Wizard Tulu Dolphin 70B](
|
|
|
1425
1880
|
"functionCall": false,
|
|
1426
1881
|
"id": "sophosympatheia/midnight-rose-70b",
|
|
1427
1882
|
"maxTokens": undefined,
|
|
1883
|
+
"pricing": {
|
|
1884
|
+
"input": 0.8,
|
|
1885
|
+
"output": 0.8,
|
|
1886
|
+
},
|
|
1428
1887
|
"reasoning": false,
|
|
1888
|
+
"releasedAt": "2024-03-22",
|
|
1429
1889
|
"vision": false,
|
|
1430
1890
|
},
|
|
1431
1891
|
{
|
|
@@ -1437,10 +1897,15 @@ Read the launch post [here](https://txt.cohere.com/command-r/).
|
|
|
1437
1897
|
Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).",
|
|
1438
1898
|
"displayName": "Cohere: Command R",
|
|
1439
1899
|
"enabled": false,
|
|
1440
|
-
"functionCall":
|
|
1900
|
+
"functionCall": true,
|
|
1441
1901
|
"id": "cohere/command-r",
|
|
1442
1902
|
"maxTokens": 4000,
|
|
1903
|
+
"pricing": {
|
|
1904
|
+
"input": 0.5,
|
|
1905
|
+
"output": 1.5,
|
|
1906
|
+
},
|
|
1443
1907
|
"reasoning": false,
|
|
1908
|
+
"releasedAt": "2024-03-14",
|
|
1444
1909
|
"vision": false,
|
|
1445
1910
|
},
|
|
1446
1911
|
{
|
|
@@ -1453,7 +1918,12 @@ Use of this model is subject to Cohere's [Acceptable Use Policy](https://docs.co
|
|
|
1453
1918
|
"functionCall": false,
|
|
1454
1919
|
"id": "cohere/command",
|
|
1455
1920
|
"maxTokens": 4000,
|
|
1921
|
+
"pricing": {
|
|
1922
|
+
"input": 1,
|
|
1923
|
+
"output": 2,
|
|
1924
|
+
},
|
|
1456
1925
|
"reasoning": false,
|
|
1926
|
+
"releasedAt": "2024-03-14",
|
|
1457
1927
|
"vision": false,
|
|
1458
1928
|
},
|
|
1459
1929
|
{
|
|
@@ -1469,7 +1939,12 @@ See the launch announcement and benchmark results [here](https://www.anthropic.c
|
|
|
1469
1939
|
"functionCall": true,
|
|
1470
1940
|
"id": "anthropic/claude-3-haiku",
|
|
1471
1941
|
"maxTokens": 4096,
|
|
1942
|
+
"pricing": {
|
|
1943
|
+
"input": 0.25,
|
|
1944
|
+
"output": 1.25,
|
|
1945
|
+
},
|
|
1472
1946
|
"reasoning": false,
|
|
1947
|
+
"releasedAt": "2024-03-13",
|
|
1473
1948
|
"vision": true,
|
|
1474
1949
|
},
|
|
1475
1950
|
{
|
|
@@ -1484,10 +1959,15 @@ See the launch announcement and benchmark results [here](https://www.anthropic.c
|
|
|
1484
1959
|
_This is a faster endpoint, made available in collaboration with Anthropic, that is self-moderated: response moderation happens on the provider's side instead of OpenRouter's. For requests that pass moderation, it's identical to the [Standard](/models/anthropic/claude-3-haiku) variant._",
|
|
1485
1960
|
"displayName": "Anthropic: Claude 3 Haiku (self-moderated)",
|
|
1486
1961
|
"enabled": false,
|
|
1487
|
-
"functionCall":
|
|
1962
|
+
"functionCall": true,
|
|
1488
1963
|
"id": "anthropic/claude-3-haiku:beta",
|
|
1489
1964
|
"maxTokens": 4096,
|
|
1965
|
+
"pricing": {
|
|
1966
|
+
"input": 0.25,
|
|
1967
|
+
"output": 1.25,
|
|
1968
|
+
},
|
|
1490
1969
|
"reasoning": false,
|
|
1970
|
+
"releasedAt": "2024-03-13",
|
|
1491
1971
|
"vision": true,
|
|
1492
1972
|
},
|
|
1493
1973
|
{
|
|
@@ -1499,10 +1979,15 @@ See the launch announcement and benchmark results [here](https://www.anthropic.c
|
|
|
1499
1979
|
#multimodal",
|
|
1500
1980
|
"displayName": "Anthropic: Claude 3 Sonnet",
|
|
1501
1981
|
"enabled": false,
|
|
1502
|
-
"functionCall":
|
|
1982
|
+
"functionCall": true,
|
|
1503
1983
|
"id": "anthropic/claude-3-sonnet",
|
|
1504
1984
|
"maxTokens": 4096,
|
|
1985
|
+
"pricing": {
|
|
1986
|
+
"input": 3,
|
|
1987
|
+
"output": 15,
|
|
1988
|
+
},
|
|
1505
1989
|
"reasoning": false,
|
|
1990
|
+
"releasedAt": "2024-03-05",
|
|
1506
1991
|
"vision": true,
|
|
1507
1992
|
},
|
|
1508
1993
|
{
|
|
@@ -1516,10 +2001,15 @@ See the launch announcement and benchmark results [here](https://www.anthropic.c
|
|
|
1516
2001
|
_This is a faster endpoint, made available in collaboration with Anthropic, that is self-moderated: response moderation happens on the provider's side instead of OpenRouter's. For requests that pass moderation, it's identical to the [Standard](/models/anthropic/claude-3-sonnet) variant._",
|
|
1517
2002
|
"displayName": "Anthropic: Claude 3 Sonnet (self-moderated)",
|
|
1518
2003
|
"enabled": false,
|
|
1519
|
-
"functionCall":
|
|
2004
|
+
"functionCall": true,
|
|
1520
2005
|
"id": "anthropic/claude-3-sonnet:beta",
|
|
1521
2006
|
"maxTokens": 4096,
|
|
2007
|
+
"pricing": {
|
|
2008
|
+
"input": 3,
|
|
2009
|
+
"output": 15,
|
|
2010
|
+
},
|
|
1522
2011
|
"reasoning": false,
|
|
2012
|
+
"releasedAt": "2024-03-05",
|
|
1523
2013
|
"vision": true,
|
|
1524
2014
|
},
|
|
1525
2015
|
{
|
|
@@ -1534,7 +2024,12 @@ See the launch announcement and benchmark results [here](https://www.anthropic.c
|
|
|
1534
2024
|
"functionCall": true,
|
|
1535
2025
|
"id": "anthropic/claude-3-opus",
|
|
1536
2026
|
"maxTokens": 4096,
|
|
2027
|
+
"pricing": {
|
|
2028
|
+
"input": 15,
|
|
2029
|
+
"output": 75,
|
|
2030
|
+
},
|
|
1537
2031
|
"reasoning": false,
|
|
2032
|
+
"releasedAt": "2024-03-05",
|
|
1538
2033
|
"vision": true,
|
|
1539
2034
|
},
|
|
1540
2035
|
{
|
|
@@ -1548,10 +2043,15 @@ See the launch announcement and benchmark results [here](https://www.anthropic.c
|
|
|
1548
2043
|
_This is a faster endpoint, made available in collaboration with Anthropic, that is self-moderated: response moderation happens on the provider's side instead of OpenRouter's. For requests that pass moderation, it's identical to the [Standard](/models/anthropic/claude-3-opus) variant._",
|
|
1549
2044
|
"displayName": "Anthropic: Claude 3 Opus (self-moderated)",
|
|
1550
2045
|
"enabled": false,
|
|
1551
|
-
"functionCall":
|
|
2046
|
+
"functionCall": true,
|
|
1552
2047
|
"id": "anthropic/claude-3-opus:beta",
|
|
1553
2048
|
"maxTokens": 4096,
|
|
2049
|
+
"pricing": {
|
|
2050
|
+
"input": 15,
|
|
2051
|
+
"output": 75,
|
|
2052
|
+
},
|
|
1554
2053
|
"reasoning": false,
|
|
2054
|
+
"releasedAt": "2024-03-05",
|
|
1555
2055
|
"vision": true,
|
|
1556
2056
|
},
|
|
1557
2057
|
{
|
|
@@ -1561,10 +2061,15 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
1561
2061
|
It is fluent in English, French, Spanish, German, and Italian, with high grammatical accuracy, and its long context window allows precise information recall from large documents.",
|
|
1562
2062
|
"displayName": "Mistral Large",
|
|
1563
2063
|
"enabled": false,
|
|
1564
|
-
"functionCall":
|
|
2064
|
+
"functionCall": true,
|
|
1565
2065
|
"id": "mistralai/mistral-large",
|
|
1566
2066
|
"maxTokens": undefined,
|
|
2067
|
+
"pricing": {
|
|
2068
|
+
"input": 3,
|
|
2069
|
+
"output": 9,
|
|
2070
|
+
},
|
|
1567
2071
|
"reasoning": false,
|
|
2072
|
+
"releasedAt": "2024-02-26",
|
|
1568
2073
|
"vision": false,
|
|
1569
2074
|
},
|
|
1570
2075
|
{
|
|
@@ -1577,7 +2082,12 @@ It is fluent in English, French, Spanish, German, and Italian, with high grammat
|
|
|
1577
2082
|
"functionCall": true,
|
|
1578
2083
|
"id": "openai/gpt-4-turbo-preview",
|
|
1579
2084
|
"maxTokens": 4096,
|
|
2085
|
+
"pricing": {
|
|
2086
|
+
"input": 10,
|
|
2087
|
+
"output": 30,
|
|
2088
|
+
},
|
|
1580
2089
|
"reasoning": false,
|
|
2090
|
+
"releasedAt": "2024-01-25",
|
|
1581
2091
|
"vision": false,
|
|
1582
2092
|
},
|
|
1583
2093
|
{
|
|
@@ -1587,10 +2097,15 @@ It is fluent in English, French, Spanish, German, and Italian, with high grammat
|
|
|
1587
2097
|
Training data up to Sep 2021.",
|
|
1588
2098
|
"displayName": "OpenAI: GPT-3.5 Turbo (older v0613)",
|
|
1589
2099
|
"enabled": false,
|
|
1590
|
-
"functionCall":
|
|
2100
|
+
"functionCall": true,
|
|
1591
2101
|
"id": "openai/gpt-3.5-turbo-0613",
|
|
1592
2102
|
"maxTokens": 4096,
|
|
2103
|
+
"pricing": {
|
|
2104
|
+
"input": 1,
|
|
2105
|
+
"output": 2,
|
|
2106
|
+
},
|
|
1593
2107
|
"reasoning": false,
|
|
2108
|
+
"releasedAt": "2024-01-25",
|
|
1594
2109
|
"vision": false,
|
|
1595
2110
|
},
|
|
1596
2111
|
{
|
|
@@ -1605,7 +2120,12 @@ The model was trained on over 1,000,000 entries of primarily [GPT-4](/models/ope
|
|
|
1605
2120
|
"functionCall": false,
|
|
1606
2121
|
"id": "nousresearch/nous-hermes-2-mixtral-8x7b-dpo",
|
|
1607
2122
|
"maxTokens": undefined,
|
|
2123
|
+
"pricing": {
|
|
2124
|
+
"input": 0.45,
|
|
2125
|
+
"output": 0.45,
|
|
2126
|
+
},
|
|
1608
2127
|
"reasoning": false,
|
|
2128
|
+
"releasedAt": "2024-01-16",
|
|
1609
2129
|
"vision": false,
|
|
1610
2130
|
},
|
|
1611
2131
|
{
|
|
@@ -1613,10 +2133,15 @@ The model was trained on over 1,000,000 entries of primarily [GPT-4](/models/ope
|
|
|
1613
2133
|
"description": "This is Mistral AI's closed-source, medium-sided model. It's powered by a closed-source prototype and excels at reasoning, code, JSON, chat, and more. In benchmarks, it compares with many of the flagship models of other companies.",
|
|
1614
2134
|
"displayName": "Mistral Medium",
|
|
1615
2135
|
"enabled": false,
|
|
1616
|
-
"functionCall":
|
|
2136
|
+
"functionCall": true,
|
|
1617
2137
|
"id": "mistralai/mistral-medium",
|
|
1618
2138
|
"maxTokens": undefined,
|
|
2139
|
+
"pricing": {
|
|
2140
|
+
"input": 2.7,
|
|
2141
|
+
"output": 8.1,
|
|
2142
|
+
},
|
|
1619
2143
|
"reasoning": false,
|
|
2144
|
+
"releasedAt": "2024-01-10",
|
|
1620
2145
|
"vision": false,
|
|
1621
2146
|
},
|
|
1622
2147
|
{
|
|
@@ -1625,10 +2150,15 @@ The model was trained on over 1,000,000 entries of primarily [GPT-4](/models/ope
|
|
|
1625
2150
|
#moe",
|
|
1626
2151
|
"displayName": "Mistral Small",
|
|
1627
2152
|
"enabled": false,
|
|
1628
|
-
"functionCall":
|
|
2153
|
+
"functionCall": true,
|
|
1629
2154
|
"id": "mistralai/mistral-small",
|
|
1630
2155
|
"maxTokens": undefined,
|
|
2156
|
+
"pricing": {
|
|
2157
|
+
"input": 2,
|
|
2158
|
+
"output": 6,
|
|
2159
|
+
},
|
|
1631
2160
|
"reasoning": false,
|
|
2161
|
+
"releasedAt": "2024-01-10",
|
|
1632
2162
|
"vision": false,
|
|
1633
2163
|
},
|
|
1634
2164
|
{
|
|
@@ -1636,10 +2166,15 @@ The model was trained on over 1,000,000 entries of primarily [GPT-4](/models/ope
|
|
|
1636
2166
|
"description": "This model is currently powered by Mistral-7B-v0.2, and incorporates a "better" fine-tuning than [Mistral 7B](/models/mistralai/mistral-7b-instruct-v0.1), inspired by community work. It's best used for large batch processing tasks where cost is a significant factor but reasoning capabilities are not crucial.",
|
|
1637
2167
|
"displayName": "Mistral Tiny",
|
|
1638
2168
|
"enabled": false,
|
|
1639
|
-
"functionCall":
|
|
2169
|
+
"functionCall": true,
|
|
1640
2170
|
"id": "mistralai/mistral-tiny",
|
|
1641
2171
|
"maxTokens": undefined,
|
|
2172
|
+
"pricing": {
|
|
2173
|
+
"input": 0.25,
|
|
2174
|
+
"output": 0.25,
|
|
2175
|
+
},
|
|
1642
2176
|
"reasoning": false,
|
|
2177
|
+
"releasedAt": "2024-01-10",
|
|
1643
2178
|
"vision": false,
|
|
1644
2179
|
},
|
|
1645
2180
|
{
|
|
@@ -1650,7 +2185,12 @@ The model was trained on over 1,000,000 entries of primarily [GPT-4](/models/ope
|
|
|
1650
2185
|
"functionCall": false,
|
|
1651
2186
|
"id": "austism/chronos-hermes-13b",
|
|
1652
2187
|
"maxTokens": undefined,
|
|
2188
|
+
"pricing": {
|
|
2189
|
+
"input": 0.13,
|
|
2190
|
+
"output": 0.13,
|
|
2191
|
+
},
|
|
1653
2192
|
"reasoning": false,
|
|
2193
|
+
"releasedAt": "2024-01-05",
|
|
1654
2194
|
"vision": false,
|
|
1655
2195
|
},
|
|
1656
2196
|
{
|
|
@@ -1663,7 +2203,12 @@ Nous-Hermes 2 on Yi 34B outperforms all Nous-Hermes & Open-Hermes models of the
|
|
|
1663
2203
|
"functionCall": false,
|
|
1664
2204
|
"id": "nousresearch/nous-hermes-yi-34b",
|
|
1665
2205
|
"maxTokens": undefined,
|
|
2206
|
+
"pricing": {
|
|
2207
|
+
"input": 0.72,
|
|
2208
|
+
"output": 0.72,
|
|
2209
|
+
},
|
|
1666
2210
|
"reasoning": false,
|
|
2211
|
+
"releasedAt": "2024-01-02",
|
|
1667
2212
|
"vision": false,
|
|
1668
2213
|
},
|
|
1669
2214
|
{
|
|
@@ -1680,7 +2225,12 @@ An improved version of [Mistral 7B Instruct](/modelsmistralai/mistral-7b-instruc
|
|
|
1680
2225
|
"functionCall": false,
|
|
1681
2226
|
"id": "mistralai/mistral-7b-instruct-v0.2",
|
|
1682
2227
|
"maxTokens": undefined,
|
|
2228
|
+
"pricing": {
|
|
2229
|
+
"input": 0.055,
|
|
2230
|
+
"output": 0.055,
|
|
2231
|
+
},
|
|
1683
2232
|
"reasoning": false,
|
|
2233
|
+
"releasedAt": "2023-12-28",
|
|
1684
2234
|
"vision": false,
|
|
1685
2235
|
},
|
|
1686
2236
|
{
|
|
@@ -1695,7 +2245,12 @@ The model is uncensored and is stripped of alignment and bias. It requires an ex
|
|
|
1695
2245
|
"functionCall": false,
|
|
1696
2246
|
"id": "cognitivecomputations/dolphin-mixtral-8x7b",
|
|
1697
2247
|
"maxTokens": undefined,
|
|
2248
|
+
"pricing": {
|
|
2249
|
+
"input": 0.5,
|
|
2250
|
+
"output": 0.5,
|
|
2251
|
+
},
|
|
1698
2252
|
"reasoning": false,
|
|
2253
|
+
"releasedAt": "2023-12-21",
|
|
1699
2254
|
"vision": false,
|
|
1700
2255
|
},
|
|
1701
2256
|
{
|
|
@@ -1709,10 +2264,15 @@ Usage of Gemini is subject to Google's [Gemini Terms of Use](https://ai.google.d
|
|
|
1709
2264
|
#multimodal",
|
|
1710
2265
|
"displayName": "Google: Gemini Pro Vision 1.0",
|
|
1711
2266
|
"enabled": false,
|
|
1712
|
-
"functionCall":
|
|
2267
|
+
"functionCall": true,
|
|
1713
2268
|
"id": "google/gemini-pro-vision",
|
|
1714
2269
|
"maxTokens": 8192,
|
|
2270
|
+
"pricing": {
|
|
2271
|
+
"input": 0.125,
|
|
2272
|
+
"output": 0.375,
|
|
2273
|
+
},
|
|
1715
2274
|
"reasoning": false,
|
|
2275
|
+
"releasedAt": "2023-12-13",
|
|
1716
2276
|
"vision": true,
|
|
1717
2277
|
},
|
|
1718
2278
|
{
|
|
@@ -1724,10 +2284,15 @@ See the benchmarks and prompting guidelines from [Deepmind](https://deepmind.goo
|
|
|
1724
2284
|
Usage of Gemini is subject to Google's [Gemini Terms of Use](https://ai.google.dev/terms).",
|
|
1725
2285
|
"displayName": "Google: Gemini Pro 1.0",
|
|
1726
2286
|
"enabled": false,
|
|
1727
|
-
"functionCall":
|
|
2287
|
+
"functionCall": true,
|
|
1728
2288
|
"id": "google/gemini-pro",
|
|
1729
2289
|
"maxTokens": 32768,
|
|
2290
|
+
"pricing": {
|
|
2291
|
+
"input": 0.125,
|
|
2292
|
+
"output": 0.375,
|
|
2293
|
+
},
|
|
1730
2294
|
"reasoning": false,
|
|
2295
|
+
"releasedAt": "2023-12-13",
|
|
1731
2296
|
"vision": false,
|
|
1732
2297
|
},
|
|
1733
2298
|
{
|
|
@@ -1737,10 +2302,15 @@ Usage of Gemini is subject to Google's [Gemini Terms of Use](https://ai.google.d
|
|
|
1737
2302
|
Instruct model fine-tuned by Mistral. #moe",
|
|
1738
2303
|
"displayName": "Mixtral 8x7B Instruct",
|
|
1739
2304
|
"enabled": false,
|
|
1740
|
-
"functionCall":
|
|
2305
|
+
"functionCall": true,
|
|
1741
2306
|
"id": "mistralai/mixtral-8x7b-instruct",
|
|
1742
2307
|
"maxTokens": undefined,
|
|
2308
|
+
"pricing": {
|
|
2309
|
+
"input": 0.24,
|
|
2310
|
+
"output": 0.24,
|
|
2311
|
+
},
|
|
1743
2312
|
"reasoning": false,
|
|
2313
|
+
"releasedAt": "2023-12-10",
|
|
1744
2314
|
"vision": false,
|
|
1745
2315
|
},
|
|
1746
2316
|
{
|
|
@@ -1755,7 +2325,12 @@ _These are higher-throughput endpoints for [Mixtral 8x7B Instruct](/models/mistr
|
|
|
1755
2325
|
"functionCall": false,
|
|
1756
2326
|
"id": "mistralai/mixtral-8x7b-instruct:nitro",
|
|
1757
2327
|
"maxTokens": undefined,
|
|
2328
|
+
"pricing": {
|
|
2329
|
+
"input": 0.54,
|
|
2330
|
+
"output": 0.54,
|
|
2331
|
+
},
|
|
1758
2332
|
"reasoning": false,
|
|
2333
|
+
"releasedAt": "2023-12-10",
|
|
1759
2334
|
"vision": false,
|
|
1760
2335
|
},
|
|
1761
2336
|
{
|
|
@@ -1768,7 +2343,12 @@ _These are higher-throughput endpoints for [Mixtral 8x7B Instruct](/models/mistr
|
|
|
1768
2343
|
"functionCall": false,
|
|
1769
2344
|
"id": "mistralai/mixtral-8x7b",
|
|
1770
2345
|
"maxTokens": undefined,
|
|
2346
|
+
"pricing": {
|
|
2347
|
+
"input": 0.54,
|
|
2348
|
+
"output": 0.54,
|
|
2349
|
+
},
|
|
1771
2350
|
"reasoning": false,
|
|
2351
|
+
"releasedAt": "2023-12-10",
|
|
1772
2352
|
"vision": false,
|
|
1773
2353
|
},
|
|
1774
2354
|
{
|
|
@@ -1781,7 +2361,12 @@ StripedHyena uses a new architecture that competes with traditional Transformers
|
|
|
1781
2361
|
"functionCall": false,
|
|
1782
2362
|
"id": "togethercomputer/stripedhyena-nous-7b",
|
|
1783
2363
|
"maxTokens": undefined,
|
|
2364
|
+
"pricing": {
|
|
2365
|
+
"input": 0.18,
|
|
2366
|
+
"output": 0.18,
|
|
2367
|
+
},
|
|
1784
2368
|
"reasoning": false,
|
|
2369
|
+
"releasedAt": "2023-12-09",
|
|
1785
2370
|
"vision": false,
|
|
1786
2371
|
},
|
|
1787
2372
|
{
|
|
@@ -1798,7 +2383,12 @@ _These are free, rate-limited endpoints for [MythoMist 7B](/models/gryphe/mythom
|
|
|
1798
2383
|
"functionCall": false,
|
|
1799
2384
|
"id": "gryphe/mythomist-7b:free",
|
|
1800
2385
|
"maxTokens": 4096,
|
|
2386
|
+
"pricing": {
|
|
2387
|
+
"input": 0,
|
|
2388
|
+
"output": 0,
|
|
2389
|
+
},
|
|
1801
2390
|
"reasoning": false,
|
|
2391
|
+
"releasedAt": "2023-12-07",
|
|
1802
2392
|
"vision": false,
|
|
1803
2393
|
},
|
|
1804
2394
|
{
|
|
@@ -1813,7 +2403,12 @@ It combines [Neural Chat 7B](/models/intel/neural-chat-7b), Airoboros 7b, [Toppy
|
|
|
1813
2403
|
"functionCall": false,
|
|
1814
2404
|
"id": "gryphe/mythomist-7b",
|
|
1815
2405
|
"maxTokens": 2048,
|
|
2406
|
+
"pricing": {
|
|
2407
|
+
"input": 0.375,
|
|
2408
|
+
"output": 0.375,
|
|
2409
|
+
},
|
|
1816
2410
|
"reasoning": false,
|
|
2411
|
+
"releasedAt": "2023-12-07",
|
|
1817
2412
|
"vision": false,
|
|
1818
2413
|
},
|
|
1819
2414
|
{
|
|
@@ -1831,7 +2426,12 @@ _These are free, rate-limited endpoints for [OpenChat 3.5 7B](/models/openchat/o
|
|
|
1831
2426
|
"functionCall": false,
|
|
1832
2427
|
"id": "openchat/openchat-7b:free",
|
|
1833
2428
|
"maxTokens": 4096,
|
|
2429
|
+
"pricing": {
|
|
2430
|
+
"input": 0,
|
|
2431
|
+
"output": 0,
|
|
2432
|
+
},
|
|
1834
2433
|
"reasoning": false,
|
|
2434
|
+
"releasedAt": "2023-11-28",
|
|
1835
2435
|
"vision": false,
|
|
1836
2436
|
},
|
|
1837
2437
|
{
|
|
@@ -1847,7 +2447,12 @@ _These are free, rate-limited endpoints for [OpenChat 3.5 7B](/models/openchat/o
|
|
|
1847
2447
|
"functionCall": false,
|
|
1848
2448
|
"id": "openchat/openchat-7b",
|
|
1849
2449
|
"maxTokens": undefined,
|
|
2450
|
+
"pricing": {
|
|
2451
|
+
"input": 0.055,
|
|
2452
|
+
"output": 0.055,
|
|
2453
|
+
},
|
|
1850
2454
|
"reasoning": false,
|
|
2455
|
+
"releasedAt": "2023-11-28",
|
|
1851
2456
|
"vision": false,
|
|
1852
2457
|
},
|
|
1853
2458
|
{
|
|
@@ -1860,7 +2465,12 @@ _These are free, rate-limited endpoints for [OpenChat 3.5 7B](/models/openchat/o
|
|
|
1860
2465
|
"functionCall": false,
|
|
1861
2466
|
"id": "neversleep/noromaid-20b",
|
|
1862
2467
|
"maxTokens": 2048,
|
|
2468
|
+
"pricing": {
|
|
2469
|
+
"input": 1.5,
|
|
2470
|
+
"output": 2.25,
|
|
2471
|
+
},
|
|
1863
2472
|
"reasoning": false,
|
|
2473
|
+
"releasedAt": "2023-11-26",
|
|
1864
2474
|
"vision": false,
|
|
1865
2475
|
},
|
|
1866
2476
|
{
|
|
@@ -1871,7 +2481,12 @@ _These are free, rate-limited endpoints for [OpenChat 3.5 7B](/models/openchat/o
|
|
|
1871
2481
|
"functionCall": false,
|
|
1872
2482
|
"id": "anthropic/claude-instant-1.1",
|
|
1873
2483
|
"maxTokens": 2048,
|
|
2484
|
+
"pricing": {
|
|
2485
|
+
"input": 0.8,
|
|
2486
|
+
"output": 2.4,
|
|
2487
|
+
},
|
|
1874
2488
|
"reasoning": false,
|
|
2489
|
+
"releasedAt": "2023-11-22",
|
|
1875
2490
|
"vision": false,
|
|
1876
2491
|
},
|
|
1877
2492
|
{
|
|
@@ -1882,7 +2497,12 @@ _These are free, rate-limited endpoints for [OpenChat 3.5 7B](/models/openchat/o
|
|
|
1882
2497
|
"functionCall": false,
|
|
1883
2498
|
"id": "anthropic/claude-2.1",
|
|
1884
2499
|
"maxTokens": 4096,
|
|
2500
|
+
"pricing": {
|
|
2501
|
+
"input": 8,
|
|
2502
|
+
"output": 24,
|
|
2503
|
+
},
|
|
1885
2504
|
"reasoning": false,
|
|
2505
|
+
"releasedAt": "2023-11-22",
|
|
1886
2506
|
"vision": false,
|
|
1887
2507
|
},
|
|
1888
2508
|
{
|
|
@@ -1895,7 +2515,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
1895
2515
|
"functionCall": false,
|
|
1896
2516
|
"id": "anthropic/claude-2.1:beta",
|
|
1897
2517
|
"maxTokens": 4096,
|
|
2518
|
+
"pricing": {
|
|
2519
|
+
"input": 8,
|
|
2520
|
+
"output": 24,
|
|
2521
|
+
},
|
|
1898
2522
|
"reasoning": false,
|
|
2523
|
+
"releasedAt": "2023-11-22",
|
|
1899
2524
|
"vision": false,
|
|
1900
2525
|
},
|
|
1901
2526
|
{
|
|
@@ -1906,7 +2531,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
1906
2531
|
"functionCall": false,
|
|
1907
2532
|
"id": "anthropic/claude-2",
|
|
1908
2533
|
"maxTokens": 4096,
|
|
2534
|
+
"pricing": {
|
|
2535
|
+
"input": 8,
|
|
2536
|
+
"output": 24,
|
|
2537
|
+
},
|
|
1909
2538
|
"reasoning": false,
|
|
2539
|
+
"releasedAt": "2023-11-22",
|
|
1910
2540
|
"vision": false,
|
|
1911
2541
|
},
|
|
1912
2542
|
{
|
|
@@ -1919,7 +2549,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
1919
2549
|
"functionCall": false,
|
|
1920
2550
|
"id": "anthropic/claude-2:beta",
|
|
1921
2551
|
"maxTokens": 4096,
|
|
2552
|
+
"pricing": {
|
|
2553
|
+
"input": 8,
|
|
2554
|
+
"output": 24,
|
|
2555
|
+
},
|
|
1922
2556
|
"reasoning": false,
|
|
2557
|
+
"releasedAt": "2023-11-22",
|
|
1923
2558
|
"vision": false,
|
|
1924
2559
|
},
|
|
1925
2560
|
{
|
|
@@ -1931,7 +2566,12 @@ Potentially the most interesting finding from training on a good ratio (est. of
|
|
|
1931
2566
|
"functionCall": false,
|
|
1932
2567
|
"id": "teknium/openhermes-2.5-mistral-7b",
|
|
1933
2568
|
"maxTokens": undefined,
|
|
2569
|
+
"pricing": {
|
|
2570
|
+
"input": 0.17,
|
|
2571
|
+
"output": 0.17,
|
|
2572
|
+
},
|
|
1934
2573
|
"reasoning": false,
|
|
2574
|
+
"releasedAt": "2023-11-20",
|
|
1935
2575
|
"vision": false,
|
|
1936
2576
|
},
|
|
1937
2577
|
{
|
|
@@ -1946,7 +2586,12 @@ Potentially the most interesting finding from training on a good ratio (est. of
|
|
|
1946
2586
|
"functionCall": false,
|
|
1947
2587
|
"id": "openai/gpt-4-vision-preview",
|
|
1948
2588
|
"maxTokens": 4096,
|
|
2589
|
+
"pricing": {
|
|
2590
|
+
"input": 10,
|
|
2591
|
+
"output": 30,
|
|
2592
|
+
},
|
|
1949
2593
|
"reasoning": false,
|
|
2594
|
+
"releasedAt": "2023-11-13",
|
|
1950
2595
|
"vision": true,
|
|
1951
2596
|
},
|
|
1952
2597
|
{
|
|
@@ -1960,7 +2605,12 @@ A multi-model merge of several LLaMA2 70B finetunes for roleplaying and creative
|
|
|
1960
2605
|
"functionCall": false,
|
|
1961
2606
|
"id": "lizpreciatior/lzlv-70b-fp16-hf",
|
|
1962
2607
|
"maxTokens": undefined,
|
|
2608
|
+
"pricing": {
|
|
2609
|
+
"input": 0.35,
|
|
2610
|
+
"output": 0.4,
|
|
2611
|
+
},
|
|
1963
2612
|
"reasoning": false,
|
|
2613
|
+
"releasedAt": "2023-11-12",
|
|
1964
2614
|
"vision": false,
|
|
1965
2615
|
},
|
|
1966
2616
|
{
|
|
@@ -1977,7 +2627,12 @@ Credits to
|
|
|
1977
2627
|
"functionCall": false,
|
|
1978
2628
|
"id": "alpindale/goliath-120b",
|
|
1979
2629
|
"maxTokens": 400,
|
|
2630
|
+
"pricing": {
|
|
2631
|
+
"input": 9.375,
|
|
2632
|
+
"output": 9.375,
|
|
2633
|
+
},
|
|
1980
2634
|
"reasoning": false,
|
|
2635
|
+
"releasedAt": "2023-11-10",
|
|
1981
2636
|
"vision": false,
|
|
1982
2637
|
},
|
|
1983
2638
|
{
|
|
@@ -1998,7 +2653,12 @@ _These are free, rate-limited endpoints for [Toppy M 7B](/models/undi95/toppy-m-
|
|
|
1998
2653
|
"functionCall": false,
|
|
1999
2654
|
"id": "undi95/toppy-m-7b:free",
|
|
2000
2655
|
"maxTokens": 2048,
|
|
2656
|
+
"pricing": {
|
|
2657
|
+
"input": 0,
|
|
2658
|
+
"output": 0,
|
|
2659
|
+
},
|
|
2001
2660
|
"reasoning": false,
|
|
2661
|
+
"releasedAt": "2023-11-10",
|
|
2002
2662
|
"vision": false,
|
|
2003
2663
|
},
|
|
2004
2664
|
{
|
|
@@ -2017,7 +2677,12 @@ List of merged models:
|
|
|
2017
2677
|
"functionCall": false,
|
|
2018
2678
|
"id": "undi95/toppy-m-7b",
|
|
2019
2679
|
"maxTokens": undefined,
|
|
2680
|
+
"pricing": {
|
|
2681
|
+
"input": 0.07,
|
|
2682
|
+
"output": 0.07,
|
|
2683
|
+
},
|
|
2020
2684
|
"reasoning": false,
|
|
2685
|
+
"releasedAt": "2023-11-10",
|
|
2021
2686
|
"vision": false,
|
|
2022
2687
|
},
|
|
2023
2688
|
{
|
|
@@ -2038,7 +2703,12 @@ _These are higher-throughput endpoints for [Toppy M 7B](/models/undi95/toppy-m-7
|
|
|
2038
2703
|
"functionCall": false,
|
|
2039
2704
|
"id": "undi95/toppy-m-7b:nitro",
|
|
2040
2705
|
"maxTokens": undefined,
|
|
2706
|
+
"pricing": {
|
|
2707
|
+
"input": 0.07,
|
|
2708
|
+
"output": 0.07,
|
|
2709
|
+
},
|
|
2041
2710
|
"reasoning": false,
|
|
2711
|
+
"releasedAt": "2023-11-10",
|
|
2042
2712
|
"vision": false,
|
|
2043
2713
|
},
|
|
2044
2714
|
{
|
|
@@ -2051,7 +2721,12 @@ A major redesign of this router is coming soon. Stay tuned on [Discord](https://
|
|
|
2051
2721
|
"functionCall": false,
|
|
2052
2722
|
"id": "openrouter/auto",
|
|
2053
2723
|
"maxTokens": undefined,
|
|
2724
|
+
"pricing": {
|
|
2725
|
+
"input": undefined,
|
|
2726
|
+
"output": undefined,
|
|
2727
|
+
},
|
|
2054
2728
|
"reasoning": false,
|
|
2729
|
+
"releasedAt": "2023-11-08",
|
|
2055
2730
|
"vision": false,
|
|
2056
2731
|
},
|
|
2057
2732
|
{
|
|
@@ -2064,8 +2739,13 @@ Training data: up to April 2023.",
|
|
|
2064
2739
|
"functionCall": true,
|
|
2065
2740
|
"id": "openai/gpt-4-1106-preview",
|
|
2066
2741
|
"maxTokens": 4096,
|
|
2742
|
+
"pricing": {
|
|
2743
|
+
"input": 10,
|
|
2744
|
+
"output": 30,
|
|
2745
|
+
},
|
|
2067
2746
|
"reasoning": false,
|
|
2068
|
-
"
|
|
2747
|
+
"releasedAt": "2023-11-06",
|
|
2748
|
+
"vision": false,
|
|
2069
2749
|
},
|
|
2070
2750
|
{
|
|
2071
2751
|
"contextWindowTokens": 16385,
|
|
@@ -2075,7 +2755,12 @@ Training data: up to April 2023.",
|
|
|
2075
2755
|
"functionCall": true,
|
|
2076
2756
|
"id": "openai/gpt-3.5-turbo-1106",
|
|
2077
2757
|
"maxTokens": 4096,
|
|
2758
|
+
"pricing": {
|
|
2759
|
+
"input": 1,
|
|
2760
|
+
"output": 2,
|
|
2761
|
+
},
|
|
2078
2762
|
"reasoning": false,
|
|
2763
|
+
"releasedAt": "2023-11-06",
|
|
2079
2764
|
"vision": false,
|
|
2080
2765
|
},
|
|
2081
2766
|
{
|
|
@@ -2086,7 +2771,12 @@ Training data: up to April 2023.",
|
|
|
2086
2771
|
"functionCall": false,
|
|
2087
2772
|
"id": "google/palm-2-codechat-bison-32k",
|
|
2088
2773
|
"maxTokens": 32768,
|
|
2774
|
+
"pricing": {
|
|
2775
|
+
"input": 0.25,
|
|
2776
|
+
"output": 0.5,
|
|
2777
|
+
},
|
|
2089
2778
|
"reasoning": false,
|
|
2779
|
+
"releasedAt": "2023-11-03",
|
|
2090
2780
|
"vision": false,
|
|
2091
2781
|
},
|
|
2092
2782
|
{
|
|
@@ -2097,7 +2787,12 @@ Training data: up to April 2023.",
|
|
|
2097
2787
|
"functionCall": false,
|
|
2098
2788
|
"id": "google/palm-2-chat-bison-32k",
|
|
2099
2789
|
"maxTokens": 32768,
|
|
2790
|
+
"pricing": {
|
|
2791
|
+
"input": 0.25,
|
|
2792
|
+
"output": 0.5,
|
|
2793
|
+
},
|
|
2100
2794
|
"reasoning": false,
|
|
2795
|
+
"releasedAt": "2023-11-03",
|
|
2101
2796
|
"vision": false,
|
|
2102
2797
|
},
|
|
2103
2798
|
{
|
|
@@ -2110,7 +2805,12 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2110
2805
|
"functionCall": false,
|
|
2111
2806
|
"id": "jondurbin/airoboros-l2-70b",
|
|
2112
2807
|
"maxTokens": undefined,
|
|
2808
|
+
"pricing": {
|
|
2809
|
+
"input": 0.5,
|
|
2810
|
+
"output": 0.5,
|
|
2811
|
+
},
|
|
2113
2812
|
"reasoning": false,
|
|
2813
|
+
"releasedAt": "2023-10-29",
|
|
2114
2814
|
"vision": false,
|
|
2115
2815
|
},
|
|
2116
2816
|
{
|
|
@@ -2121,7 +2821,12 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2121
2821
|
"functionCall": false,
|
|
2122
2822
|
"id": "xwin-lm/xwin-lm-70b",
|
|
2123
2823
|
"maxTokens": 400,
|
|
2824
|
+
"pricing": {
|
|
2825
|
+
"input": 3.75,
|
|
2826
|
+
"output": 3.75,
|
|
2827
|
+
},
|
|
2124
2828
|
"reasoning": false,
|
|
2829
|
+
"releasedAt": "2023-10-15",
|
|
2125
2830
|
"vision": false,
|
|
2126
2831
|
},
|
|
2127
2832
|
{
|
|
@@ -2132,7 +2837,12 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2132
2837
|
"functionCall": true,
|
|
2133
2838
|
"id": "mistralai/mistral-7b-instruct-v0.1",
|
|
2134
2839
|
"maxTokens": undefined,
|
|
2840
|
+
"pricing": {
|
|
2841
|
+
"input": 0.055,
|
|
2842
|
+
"output": 0.055,
|
|
2843
|
+
},
|
|
2135
2844
|
"reasoning": false,
|
|
2845
|
+
"releasedAt": "2023-09-28",
|
|
2136
2846
|
"vision": false,
|
|
2137
2847
|
},
|
|
2138
2848
|
{
|
|
@@ -2143,7 +2853,12 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2143
2853
|
"functionCall": false,
|
|
2144
2854
|
"id": "openai/gpt-3.5-turbo-instruct",
|
|
2145
2855
|
"maxTokens": 4096,
|
|
2856
|
+
"pricing": {
|
|
2857
|
+
"input": 1.5,
|
|
2858
|
+
"output": 2,
|
|
2859
|
+
},
|
|
2146
2860
|
"reasoning": false,
|
|
2861
|
+
"releasedAt": "2023-09-28",
|
|
2147
2862
|
"vision": false,
|
|
2148
2863
|
},
|
|
2149
2864
|
{
|
|
@@ -2154,7 +2869,12 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2154
2869
|
"functionCall": false,
|
|
2155
2870
|
"id": "pygmalionai/mythalion-13b",
|
|
2156
2871
|
"maxTokens": 400,
|
|
2872
|
+
"pricing": {
|
|
2873
|
+
"input": 1.125,
|
|
2874
|
+
"output": 1.125,
|
|
2875
|
+
},
|
|
2157
2876
|
"reasoning": false,
|
|
2877
|
+
"releasedAt": "2023-09-02",
|
|
2158
2878
|
"vision": false,
|
|
2159
2879
|
},
|
|
2160
2880
|
{
|
|
@@ -2162,10 +2882,15 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2162
2882
|
"description": "GPT-4-32k is an extended version of GPT-4, with the same capabilities but quadrupled context length, allowing for processing up to 40 pages of text in a single pass. This is particularly beneficial for handling longer content like interacting with PDFs without an external vector database. Training data: up to Sep 2021.",
|
|
2163
2883
|
"displayName": "OpenAI: GPT-4 32k (older v0314)",
|
|
2164
2884
|
"enabled": false,
|
|
2165
|
-
"functionCall":
|
|
2885
|
+
"functionCall": true,
|
|
2166
2886
|
"id": "openai/gpt-4-32k-0314",
|
|
2167
2887
|
"maxTokens": 4096,
|
|
2888
|
+
"pricing": {
|
|
2889
|
+
"input": 60,
|
|
2890
|
+
"output": 120,
|
|
2891
|
+
},
|
|
2168
2892
|
"reasoning": false,
|
|
2893
|
+
"releasedAt": "2023-08-28",
|
|
2169
2894
|
"vision": false,
|
|
2170
2895
|
},
|
|
2171
2896
|
{
|
|
@@ -2173,10 +2898,15 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2173
2898
|
"description": "GPT-4-32k is an extended version of GPT-4, with the same capabilities but quadrupled context length, allowing for processing up to 40 pages of text in a single pass. This is particularly beneficial for handling longer content like interacting with PDFs without an external vector database. Training data: up to Sep 2021.",
|
|
2174
2899
|
"displayName": "OpenAI: GPT-4 32k",
|
|
2175
2900
|
"enabled": false,
|
|
2176
|
-
"functionCall":
|
|
2901
|
+
"functionCall": true,
|
|
2177
2902
|
"id": "openai/gpt-4-32k",
|
|
2178
2903
|
"maxTokens": 4096,
|
|
2904
|
+
"pricing": {
|
|
2905
|
+
"input": 60,
|
|
2906
|
+
"output": 120,
|
|
2907
|
+
},
|
|
2179
2908
|
"reasoning": false,
|
|
2909
|
+
"releasedAt": "2023-08-28",
|
|
2180
2910
|
"vision": false,
|
|
2181
2911
|
},
|
|
2182
2912
|
{
|
|
@@ -2184,10 +2914,15 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2184
2914
|
"description": "This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost. Training data: up to Sep 2021.",
|
|
2185
2915
|
"displayName": "OpenAI: GPT-3.5 Turbo 16k",
|
|
2186
2916
|
"enabled": false,
|
|
2187
|
-
"functionCall":
|
|
2917
|
+
"functionCall": true,
|
|
2188
2918
|
"id": "openai/gpt-3.5-turbo-16k",
|
|
2189
2919
|
"maxTokens": 4096,
|
|
2920
|
+
"pricing": {
|
|
2921
|
+
"input": 3,
|
|
2922
|
+
"output": 4,
|
|
2923
|
+
},
|
|
2190
2924
|
"reasoning": false,
|
|
2925
|
+
"releasedAt": "2023-08-28",
|
|
2191
2926
|
"vision": false,
|
|
2192
2927
|
},
|
|
2193
2928
|
{
|
|
@@ -2198,7 +2933,12 @@ Currently based on [jondurbin/airoboros-l2-70b](https://huggingface.co/jondurbin
|
|
|
2198
2933
|
"functionCall": false,
|
|
2199
2934
|
"id": "nousresearch/nous-hermes-llama2-13b",
|
|
2200
2935
|
"maxTokens": undefined,
|
|
2936
|
+
"pricing": {
|
|
2937
|
+
"input": 0.17,
|
|
2938
|
+
"output": 0.17,
|
|
2939
|
+
},
|
|
2201
2940
|
"reasoning": false,
|
|
2941
|
+
"releasedAt": "2023-08-20",
|
|
2202
2942
|
"vision": false,
|
|
2203
2943
|
},
|
|
2204
2944
|
{
|
|
@@ -2211,7 +2951,12 @@ _These are free, rate-limited endpoints for [Zephyr 7B](/models/huggingfaceh4/ze
|
|
|
2211
2951
|
"functionCall": false,
|
|
2212
2952
|
"id": "huggingfaceh4/zephyr-7b-beta:free",
|
|
2213
2953
|
"maxTokens": 2048,
|
|
2954
|
+
"pricing": {
|
|
2955
|
+
"input": 0,
|
|
2956
|
+
"output": 0,
|
|
2957
|
+
},
|
|
2214
2958
|
"reasoning": false,
|
|
2959
|
+
"releasedAt": "2023-08-02",
|
|
2215
2960
|
"vision": false,
|
|
2216
2961
|
},
|
|
2217
2962
|
{
|
|
@@ -2222,7 +2967,12 @@ _These are free, rate-limited endpoints for [Zephyr 7B](/models/huggingfaceh4/ze
|
|
|
2222
2967
|
"functionCall": false,
|
|
2223
2968
|
"id": "mancer/weaver",
|
|
2224
2969
|
"maxTokens": 1000,
|
|
2970
|
+
"pricing": {
|
|
2971
|
+
"input": 1.875,
|
|
2972
|
+
"output": 2.25,
|
|
2973
|
+
},
|
|
2225
2974
|
"reasoning": false,
|
|
2975
|
+
"releasedAt": "2023-08-02",
|
|
2226
2976
|
"vision": false,
|
|
2227
2977
|
},
|
|
2228
2978
|
{
|
|
@@ -2233,7 +2983,12 @@ _These are free, rate-limited endpoints for [Zephyr 7B](/models/huggingfaceh4/ze
|
|
|
2233
2983
|
"functionCall": false,
|
|
2234
2984
|
"id": "anthropic/claude-instant-1.0",
|
|
2235
2985
|
"maxTokens": 4096,
|
|
2986
|
+
"pricing": {
|
|
2987
|
+
"input": 0.8,
|
|
2988
|
+
"output": 2.4,
|
|
2989
|
+
},
|
|
2236
2990
|
"reasoning": false,
|
|
2991
|
+
"releasedAt": "2023-07-28",
|
|
2237
2992
|
"vision": false,
|
|
2238
2993
|
},
|
|
2239
2994
|
{
|
|
@@ -2244,7 +2999,12 @@ _These are free, rate-limited endpoints for [Zephyr 7B](/models/huggingfaceh4/ze
|
|
|
2244
2999
|
"functionCall": false,
|
|
2245
3000
|
"id": "anthropic/claude-1.2",
|
|
2246
3001
|
"maxTokens": 4096,
|
|
3002
|
+
"pricing": {
|
|
3003
|
+
"input": 8,
|
|
3004
|
+
"output": 24,
|
|
3005
|
+
},
|
|
2247
3006
|
"reasoning": false,
|
|
3007
|
+
"releasedAt": "2023-07-28",
|
|
2248
3008
|
"vision": false,
|
|
2249
3009
|
},
|
|
2250
3010
|
{
|
|
@@ -2255,7 +3015,12 @@ _These are free, rate-limited endpoints for [Zephyr 7B](/models/huggingfaceh4/ze
|
|
|
2255
3015
|
"functionCall": false,
|
|
2256
3016
|
"id": "anthropic/claude-1",
|
|
2257
3017
|
"maxTokens": 4096,
|
|
3018
|
+
"pricing": {
|
|
3019
|
+
"input": 8,
|
|
3020
|
+
"output": 24,
|
|
3021
|
+
},
|
|
2258
3022
|
"reasoning": false,
|
|
3023
|
+
"releasedAt": "2023-07-28",
|
|
2259
3024
|
"vision": false,
|
|
2260
3025
|
},
|
|
2261
3026
|
{
|
|
@@ -2266,7 +3031,12 @@ _These are free, rate-limited endpoints for [Zephyr 7B](/models/huggingfaceh4/ze
|
|
|
2266
3031
|
"functionCall": false,
|
|
2267
3032
|
"id": "anthropic/claude-instant-1",
|
|
2268
3033
|
"maxTokens": 4096,
|
|
3034
|
+
"pricing": {
|
|
3035
|
+
"input": 0.8,
|
|
3036
|
+
"output": 2.4,
|
|
3037
|
+
},
|
|
2269
3038
|
"reasoning": false,
|
|
3039
|
+
"releasedAt": "2023-07-28",
|
|
2270
3040
|
"vision": false,
|
|
2271
3041
|
},
|
|
2272
3042
|
{
|
|
@@ -2279,7 +3049,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
2279
3049
|
"functionCall": false,
|
|
2280
3050
|
"id": "anthropic/claude-instant-1:beta",
|
|
2281
3051
|
"maxTokens": 4096,
|
|
3052
|
+
"pricing": {
|
|
3053
|
+
"input": 0.8,
|
|
3054
|
+
"output": 2.4,
|
|
3055
|
+
},
|
|
2282
3056
|
"reasoning": false,
|
|
3057
|
+
"releasedAt": "2023-07-28",
|
|
2283
3058
|
"vision": false,
|
|
2284
3059
|
},
|
|
2285
3060
|
{
|
|
@@ -2290,7 +3065,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
2290
3065
|
"functionCall": false,
|
|
2291
3066
|
"id": "anthropic/claude-2.0",
|
|
2292
3067
|
"maxTokens": 4096,
|
|
3068
|
+
"pricing": {
|
|
3069
|
+
"input": 8,
|
|
3070
|
+
"output": 24,
|
|
3071
|
+
},
|
|
2293
3072
|
"reasoning": false,
|
|
3073
|
+
"releasedAt": "2023-07-28",
|
|
2294
3074
|
"vision": false,
|
|
2295
3075
|
},
|
|
2296
3076
|
{
|
|
@@ -2303,7 +3083,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
2303
3083
|
"functionCall": false,
|
|
2304
3084
|
"id": "anthropic/claude-2.0:beta",
|
|
2305
3085
|
"maxTokens": 4096,
|
|
3086
|
+
"pricing": {
|
|
3087
|
+
"input": 8,
|
|
3088
|
+
"output": 24,
|
|
3089
|
+
},
|
|
2306
3090
|
"reasoning": false,
|
|
3091
|
+
"releasedAt": "2023-07-28",
|
|
2307
3092
|
"vision": false,
|
|
2308
3093
|
},
|
|
2309
3094
|
{
|
|
@@ -2314,7 +3099,12 @@ _This is a faster endpoint, made available in collaboration with Anthropic, that
|
|
|
2314
3099
|
"functionCall": false,
|
|
2315
3100
|
"id": "undi95/remm-slerp-l2-13b",
|
|
2316
3101
|
"maxTokens": 400,
|
|
3102
|
+
"pricing": {
|
|
3103
|
+
"input": 1.125,
|
|
3104
|
+
"output": 1.125,
|
|
3105
|
+
},
|
|
2317
3106
|
"reasoning": false,
|
|
3107
|
+
"releasedAt": "2023-07-22",
|
|
2318
3108
|
"vision": false,
|
|
2319
3109
|
},
|
|
2320
3110
|
{
|
|
@@ -2327,7 +3117,12 @@ _These are extended-context endpoints for [ReMM SLERP 13B](/models/undi95/remm-s
|
|
|
2327
3117
|
"functionCall": false,
|
|
2328
3118
|
"id": "undi95/remm-slerp-l2-13b:extended",
|
|
2329
3119
|
"maxTokens": 400,
|
|
3120
|
+
"pricing": {
|
|
3121
|
+
"input": 1.125,
|
|
3122
|
+
"output": 1.125,
|
|
3123
|
+
},
|
|
2330
3124
|
"reasoning": false,
|
|
3125
|
+
"releasedAt": "2023-07-22",
|
|
2331
3126
|
"vision": false,
|
|
2332
3127
|
},
|
|
2333
3128
|
{
|
|
@@ -2338,7 +3133,12 @@ _These are extended-context endpoints for [ReMM SLERP 13B](/models/undi95/remm-s
|
|
|
2338
3133
|
"functionCall": false,
|
|
2339
3134
|
"id": "google/palm-2-codechat-bison",
|
|
2340
3135
|
"maxTokens": 4096,
|
|
3136
|
+
"pricing": {
|
|
3137
|
+
"input": 0.25,
|
|
3138
|
+
"output": 0.5,
|
|
3139
|
+
},
|
|
2341
3140
|
"reasoning": false,
|
|
3141
|
+
"releasedAt": "2023-07-20",
|
|
2342
3142
|
"vision": false,
|
|
2343
3143
|
},
|
|
2344
3144
|
{
|
|
@@ -2349,7 +3149,12 @@ _These are extended-context endpoints for [ReMM SLERP 13B](/models/undi95/remm-s
|
|
|
2349
3149
|
"functionCall": false,
|
|
2350
3150
|
"id": "google/palm-2-chat-bison",
|
|
2351
3151
|
"maxTokens": 4096,
|
|
3152
|
+
"pricing": {
|
|
3153
|
+
"input": 0.25,
|
|
3154
|
+
"output": 0.5,
|
|
3155
|
+
},
|
|
2352
3156
|
"reasoning": false,
|
|
3157
|
+
"releasedAt": "2023-07-20",
|
|
2353
3158
|
"vision": false,
|
|
2354
3159
|
},
|
|
2355
3160
|
{
|
|
@@ -2360,7 +3165,12 @@ _These are extended-context endpoints for [ReMM SLERP 13B](/models/undi95/remm-s
|
|
|
2360
3165
|
"functionCall": false,
|
|
2361
3166
|
"id": "gryphe/mythomax-l2-13b",
|
|
2362
3167
|
"maxTokens": undefined,
|
|
3168
|
+
"pricing": {
|
|
3169
|
+
"input": 0.1,
|
|
3170
|
+
"output": 0.1,
|
|
3171
|
+
},
|
|
2363
3172
|
"reasoning": false,
|
|
3173
|
+
"releasedAt": "2023-07-02",
|
|
2364
3174
|
"vision": false,
|
|
2365
3175
|
},
|
|
2366
3176
|
{
|
|
@@ -2373,7 +3183,12 @@ _These are higher-throughput endpoints for [MythoMax 13B](/models/gryphe/mythoma
|
|
|
2373
3183
|
"functionCall": false,
|
|
2374
3184
|
"id": "gryphe/mythomax-l2-13b:nitro",
|
|
2375
3185
|
"maxTokens": undefined,
|
|
3186
|
+
"pricing": {
|
|
3187
|
+
"input": 0.2,
|
|
3188
|
+
"output": 0.2,
|
|
3189
|
+
},
|
|
2376
3190
|
"reasoning": false,
|
|
3191
|
+
"releasedAt": "2023-07-02",
|
|
2377
3192
|
"vision": false,
|
|
2378
3193
|
},
|
|
2379
3194
|
{
|
|
@@ -2386,7 +3201,12 @@ _These are extended-context endpoints for [MythoMax 13B](/models/gryphe/mythomax
|
|
|
2386
3201
|
"functionCall": false,
|
|
2387
3202
|
"id": "gryphe/mythomax-l2-13b:extended",
|
|
2388
3203
|
"maxTokens": 400,
|
|
3204
|
+
"pricing": {
|
|
3205
|
+
"input": 1.125,
|
|
3206
|
+
"output": 1.125,
|
|
3207
|
+
},
|
|
2389
3208
|
"reasoning": false,
|
|
3209
|
+
"releasedAt": "2023-07-02",
|
|
2390
3210
|
"vision": false,
|
|
2391
3211
|
},
|
|
2392
3212
|
{
|
|
@@ -2397,7 +3217,12 @@ _These are extended-context endpoints for [MythoMax 13B](/models/gryphe/mythomax
|
|
|
2397
3217
|
"functionCall": false,
|
|
2398
3218
|
"id": "meta-llama/llama-2-13b-chat",
|
|
2399
3219
|
"maxTokens": undefined,
|
|
3220
|
+
"pricing": {
|
|
3221
|
+
"input": 0.27,
|
|
3222
|
+
"output": 0.27,
|
|
3223
|
+
},
|
|
2400
3224
|
"reasoning": false,
|
|
3225
|
+
"releasedAt": "2023-06-20",
|
|
2401
3226
|
"vision": false,
|
|
2402
3227
|
},
|
|
2403
3228
|
{
|
|
@@ -2405,10 +3230,15 @@ _These are extended-context endpoints for [MythoMax 13B](/models/gryphe/mythomax
|
|
|
2405
3230
|
"description": "GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14. Training data: up to Sep 2021.",
|
|
2406
3231
|
"displayName": "OpenAI: GPT-4 (older v0314)",
|
|
2407
3232
|
"enabled": false,
|
|
2408
|
-
"functionCall":
|
|
3233
|
+
"functionCall": true,
|
|
2409
3234
|
"id": "openai/gpt-4-0314",
|
|
2410
3235
|
"maxTokens": 4096,
|
|
3236
|
+
"pricing": {
|
|
3237
|
+
"input": 30,
|
|
3238
|
+
"output": 60,
|
|
3239
|
+
},
|
|
2411
3240
|
"reasoning": false,
|
|
3241
|
+
"releasedAt": "2023-05-28",
|
|
2412
3242
|
"vision": false,
|
|
2413
3243
|
},
|
|
2414
3244
|
{
|
|
@@ -2416,11 +3246,16 @@ _These are extended-context endpoints for [MythoMax 13B](/models/gryphe/mythomax
|
|
|
2416
3246
|
"description": "OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and advanced reasoning capabilities. Training data: up to Sep 2021.",
|
|
2417
3247
|
"displayName": "OpenAI: GPT-4",
|
|
2418
3248
|
"enabled": false,
|
|
2419
|
-
"functionCall":
|
|
3249
|
+
"functionCall": true,
|
|
2420
3250
|
"id": "openai/gpt-4",
|
|
2421
3251
|
"maxTokens": 4096,
|
|
3252
|
+
"pricing": {
|
|
3253
|
+
"input": 30,
|
|
3254
|
+
"output": 60,
|
|
3255
|
+
},
|
|
2422
3256
|
"reasoning": false,
|
|
2423
|
-
"
|
|
3257
|
+
"releasedAt": "2023-05-28",
|
|
3258
|
+
"vision": false,
|
|
2424
3259
|
},
|
|
2425
3260
|
{
|
|
2426
3261
|
"contextWindowTokens": 4095,
|
|
@@ -2432,7 +3267,12 @@ Training data up to Sep 2021.",
|
|
|
2432
3267
|
"functionCall": false,
|
|
2433
3268
|
"id": "openai/gpt-3.5-turbo-0301",
|
|
2434
3269
|
"maxTokens": 4096,
|
|
3270
|
+
"pricing": {
|
|
3271
|
+
"input": 1,
|
|
3272
|
+
"output": 2,
|
|
3273
|
+
},
|
|
2435
3274
|
"reasoning": false,
|
|
3275
|
+
"releasedAt": "2023-05-28",
|
|
2436
3276
|
"vision": false,
|
|
2437
3277
|
},
|
|
2438
3278
|
{
|
|
@@ -2445,7 +3285,12 @@ This version has a higher accuracy at responding in requested formats and a fix
|
|
|
2445
3285
|
"functionCall": true,
|
|
2446
3286
|
"id": "openai/gpt-3.5-turbo-0125",
|
|
2447
3287
|
"maxTokens": 4096,
|
|
3288
|
+
"pricing": {
|
|
3289
|
+
"input": 0.5,
|
|
3290
|
+
"output": 1.5,
|
|
3291
|
+
},
|
|
2448
3292
|
"reasoning": false,
|
|
3293
|
+
"releasedAt": "2023-05-28",
|
|
2449
3294
|
"vision": false,
|
|
2450
3295
|
},
|
|
2451
3296
|
{
|
|
@@ -2455,10 +3300,15 @@ This version has a higher accuracy at responding in requested formats and a fix
|
|
|
2455
3300
|
Training data up to Sep 2021.",
|
|
2456
3301
|
"displayName": "OpenAI: GPT-3.5 Turbo",
|
|
2457
3302
|
"enabled": false,
|
|
2458
|
-
"functionCall":
|
|
3303
|
+
"functionCall": true,
|
|
2459
3304
|
"id": "openai/gpt-3.5-turbo",
|
|
2460
3305
|
"maxTokens": 4096,
|
|
3306
|
+
"pricing": {
|
|
3307
|
+
"input": 0.5,
|
|
3308
|
+
"output": 1.5,
|
|
3309
|
+
},
|
|
2461
3310
|
"reasoning": false,
|
|
3311
|
+
"releasedAt": "2023-05-28",
|
|
2462
3312
|
"vision": false,
|
|
2463
3313
|
},
|
|
2464
3314
|
]
|