@mariozechner/pi-ai 0.5.44 → 0.5.45

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.
@@ -189,6 +189,142 @@ export const MODELS = {
189
189
  contextWindow: 200000,
190
190
  maxTokens: 4096,
191
191
  },
192
+ "claude-haiku-4-5-20251001": {
193
+ id: "claude-haiku-4-5-20251001",
194
+ name: "Claude Haiku 4.5",
195
+ api: "anthropic-messages",
196
+ provider: "anthropic",
197
+ baseUrl: "https://api.anthropic.com",
198
+ reasoning: true,
199
+ input: ["text", "image"],
200
+ cost: {
201
+ input: 1,
202
+ output: 5,
203
+ cacheRead: 0.1,
204
+ cacheWrite: 1.25,
205
+ },
206
+ contextWindow: 200000,
207
+ maxTokens: 64000,
208
+ },
209
+ "claude-sonnet-4-0": {
210
+ id: "claude-sonnet-4-0",
211
+ name: "Claude Sonnet 4",
212
+ api: "anthropic-messages",
213
+ provider: "anthropic",
214
+ baseUrl: "https://api.anthropic.com",
215
+ reasoning: true,
216
+ input: ["text", "image"],
217
+ cost: {
218
+ input: 3,
219
+ output: 15,
220
+ cacheRead: 0.3,
221
+ cacheWrite: 3.75,
222
+ },
223
+ contextWindow: 200000,
224
+ maxTokens: 64000,
225
+ },
226
+ "claude-3-7-sonnet-latest": {
227
+ id: "claude-3-7-sonnet-latest",
228
+ name: "Claude Sonnet 3.7",
229
+ api: "anthropic-messages",
230
+ provider: "anthropic",
231
+ baseUrl: "https://api.anthropic.com",
232
+ reasoning: true,
233
+ input: ["text", "image"],
234
+ cost: {
235
+ input: 3,
236
+ output: 15,
237
+ cacheRead: 0.3,
238
+ cacheWrite: 3.75,
239
+ },
240
+ contextWindow: 200000,
241
+ maxTokens: 64000,
242
+ },
243
+ "claude-sonnet-4-5": {
244
+ id: "claude-sonnet-4-5",
245
+ name: "Claude Sonnet 4.5",
246
+ api: "anthropic-messages",
247
+ provider: "anthropic",
248
+ baseUrl: "https://api.anthropic.com",
249
+ reasoning: true,
250
+ input: ["text", "image"],
251
+ cost: {
252
+ input: 3,
253
+ output: 15,
254
+ cacheRead: 0.3,
255
+ cacheWrite: 3.75,
256
+ },
257
+ contextWindow: 200000,
258
+ maxTokens: 64000,
259
+ },
260
+ "claude-3-5-haiku-latest": {
261
+ id: "claude-3-5-haiku-latest",
262
+ name: "Claude Haiku 3.5",
263
+ api: "anthropic-messages",
264
+ provider: "anthropic",
265
+ baseUrl: "https://api.anthropic.com",
266
+ reasoning: false,
267
+ input: ["text", "image"],
268
+ cost: {
269
+ input: 0.8,
270
+ output: 4,
271
+ cacheRead: 0.08,
272
+ cacheWrite: 1,
273
+ },
274
+ contextWindow: 200000,
275
+ maxTokens: 8192,
276
+ },
277
+ "claude-haiku-4-5": {
278
+ id: "claude-haiku-4-5",
279
+ name: "Claude Haiku 4.5",
280
+ api: "anthropic-messages",
281
+ provider: "anthropic",
282
+ baseUrl: "https://api.anthropic.com",
283
+ reasoning: true,
284
+ input: ["text", "image"],
285
+ cost: {
286
+ input: 1,
287
+ output: 5,
288
+ cacheRead: 0.1,
289
+ cacheWrite: 1.25,
290
+ },
291
+ contextWindow: 200000,
292
+ maxTokens: 64000,
293
+ },
294
+ "claude-opus-4-1": {
295
+ id: "claude-opus-4-1",
296
+ name: "Claude Opus 4.1",
297
+ api: "anthropic-messages",
298
+ provider: "anthropic",
299
+ baseUrl: "https://api.anthropic.com",
300
+ reasoning: true,
301
+ input: ["text", "image"],
302
+ cost: {
303
+ input: 15,
304
+ output: 75,
305
+ cacheRead: 1.5,
306
+ cacheWrite: 18.75,
307
+ },
308
+ contextWindow: 200000,
309
+ maxTokens: 32000,
310
+ },
311
+ "claude-opus-4-0": {
312
+ id: "claude-opus-4-0",
313
+ name: "Claude Opus 4",
314
+ api: "anthropic-messages",
315
+ provider: "anthropic",
316
+ baseUrl: "https://api.anthropic.com",
317
+ reasoning: true,
318
+ input: ["text", "image"],
319
+ cost: {
320
+ input: 15,
321
+ output: 75,
322
+ cacheRead: 1.5,
323
+ cacheWrite: 18.75,
324
+ },
325
+ contextWindow: 200000,
326
+ maxTokens: 32000,
327
+ },
192
328
  },
193
329
  google: {
194
330
  "gemini-2.5-flash-preview-05-20": {
@@ -251,9 +387,9 @@ export const MODELS = {
251
387
  reasoning: true,
252
388
  input: ["text", "image"],
253
389
  cost: {
254
- input: 0.15,
255
- output: 0.6,
256
- cacheRead: 0.0375,
390
+ input: 0.3,
391
+ output: 2.5,
392
+ cacheRead: 0.075,
257
393
  cacheWrite: 0,
258
394
  },
259
395
  contextWindow: 1048576,
@@ -361,6 +497,23 @@ export const MODELS = {
361
497
  contextWindow: 1048576,
362
498
  maxTokens: 65536,
363
499
  },
500
+ "gemini-live-2.5-flash": {
501
+ id: "gemini-live-2.5-flash",
502
+ name: "Gemini Live 2.5 Flash",
503
+ api: "google-generative-ai",
504
+ provider: "google",
505
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
506
+ reasoning: true,
507
+ input: ["text", "image"],
508
+ cost: {
509
+ input: 0.5,
510
+ output: 2,
511
+ cacheRead: 0,
512
+ cacheWrite: 0,
513
+ },
514
+ contextWindow: 128000,
515
+ maxTokens: 8000,
516
+ },
364
517
  "gemini-2.5-flash-lite-preview-06-17": {
365
518
  id: "gemini-2.5-flash-lite-preview-06-17",
366
519
  name: "Gemini 2.5 Flash Lite Preview 06-17",
@@ -387,9 +540,9 @@ export const MODELS = {
387
540
  reasoning: true,
388
541
  input: ["text", "image"],
389
542
  cost: {
390
- input: 0.15,
391
- output: 0.6,
392
- cacheRead: 0.0375,
543
+ input: 0.3,
544
+ output: 2.5,
545
+ cacheRead: 0.075,
393
546
  cacheWrite: 0,
394
547
  },
395
548
  contextWindow: 1048576,
@@ -695,9 +848,9 @@ export const MODELS = {
695
848
  reasoning: true,
696
849
  input: ["text", "image"],
697
850
  cost: {
698
- input: 0,
699
- output: 0,
700
- cacheRead: 0,
851
+ input: 1.25,
852
+ output: 10,
853
+ cacheRead: 0.125,
701
854
  cacheWrite: 0,
702
855
  },
703
856
  contextWindow: 400000,
@@ -1648,6 +1801,159 @@ export const MODELS = {
1648
1801
  },
1649
1802
  },
1650
1803
  openrouter: {
1804
+ "openrouter/andromeda-alpha": {
1805
+ id: "openrouter/andromeda-alpha",
1806
+ name: "Andromeda Alpha",
1807
+ api: "openai-completions",
1808
+ provider: "openrouter",
1809
+ baseUrl: "https://openrouter.ai/api/v1",
1810
+ reasoning: true,
1811
+ input: ["text", "image"],
1812
+ cost: {
1813
+ input: 0,
1814
+ output: 0,
1815
+ cacheRead: 0,
1816
+ cacheWrite: 0,
1817
+ },
1818
+ contextWindow: 128000,
1819
+ maxTokens: 128000,
1820
+ },
1821
+ "deepcogito/cogito-v2-preview-llama-405b": {
1822
+ id: "deepcogito/cogito-v2-preview-llama-405b",
1823
+ name: "Deep Cogito: Cogito V2 Preview Llama 405B",
1824
+ api: "openai-completions",
1825
+ provider: "openrouter",
1826
+ baseUrl: "https://openrouter.ai/api/v1",
1827
+ reasoning: true,
1828
+ input: ["text"],
1829
+ cost: {
1830
+ input: 3.5,
1831
+ output: 3.5,
1832
+ cacheRead: 0,
1833
+ cacheWrite: 0,
1834
+ },
1835
+ contextWindow: 32768,
1836
+ maxTokens: 4096,
1837
+ },
1838
+ "qwen/qwen3-vl-8b-thinking": {
1839
+ id: "qwen/qwen3-vl-8b-thinking",
1840
+ name: "Qwen: Qwen3 VL 8B Thinking",
1841
+ api: "openai-completions",
1842
+ provider: "openrouter",
1843
+ baseUrl: "https://openrouter.ai/api/v1",
1844
+ reasoning: true,
1845
+ input: ["text", "image"],
1846
+ cost: {
1847
+ input: 0.18,
1848
+ output: 2.0999999999999996,
1849
+ cacheRead: 0,
1850
+ cacheWrite: 0,
1851
+ },
1852
+ contextWindow: 256000,
1853
+ maxTokens: 32768,
1854
+ },
1855
+ "qwen/qwen3-vl-8b-instruct": {
1856
+ id: "qwen/qwen3-vl-8b-instruct",
1857
+ name: "Qwen: Qwen3 VL 8B Instruct",
1858
+ api: "openai-completions",
1859
+ provider: "openrouter",
1860
+ baseUrl: "https://openrouter.ai/api/v1",
1861
+ reasoning: false,
1862
+ input: ["text", "image"],
1863
+ cost: {
1864
+ input: 0.08,
1865
+ output: 0.5,
1866
+ cacheRead: 0,
1867
+ cacheWrite: 0,
1868
+ },
1869
+ contextWindow: 131072,
1870
+ maxTokens: 32768,
1871
+ },
1872
+ "inclusionai/ring-1t": {
1873
+ id: "inclusionai/ring-1t",
1874
+ name: "inclusionAI: Ring 1T",
1875
+ api: "openai-completions",
1876
+ provider: "openrouter",
1877
+ baseUrl: "https://openrouter.ai/api/v1",
1878
+ reasoning: true,
1879
+ input: ["text"],
1880
+ cost: {
1881
+ input: 0.5700000000000001,
1882
+ output: 2.2800000000000002,
1883
+ cacheRead: 0,
1884
+ cacheWrite: 0,
1885
+ },
1886
+ contextWindow: 131072,
1887
+ maxTokens: 131072,
1888
+ },
1889
+ "inclusionai/ling-1t": {
1890
+ id: "inclusionai/ling-1t",
1891
+ name: "inclusionAI: Ling-1T",
1892
+ api: "openai-completions",
1893
+ provider: "openrouter",
1894
+ baseUrl: "https://openrouter.ai/api/v1",
1895
+ reasoning: false,
1896
+ input: ["text"],
1897
+ cost: {
1898
+ input: 0.39999999999999997,
1899
+ output: 2,
1900
+ cacheRead: 0,
1901
+ cacheWrite: 0,
1902
+ },
1903
+ contextWindow: 131072,
1904
+ maxTokens: 131072,
1905
+ },
1906
+ "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
1907
+ id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
1908
+ name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
1909
+ api: "openai-completions",
1910
+ provider: "openrouter",
1911
+ baseUrl: "https://openrouter.ai/api/v1",
1912
+ reasoning: true,
1913
+ input: ["text"],
1914
+ cost: {
1915
+ input: 0.09999999999999999,
1916
+ output: 0.39999999999999997,
1917
+ cacheRead: 0,
1918
+ cacheWrite: 0,
1919
+ },
1920
+ contextWindow: 131072,
1921
+ maxTokens: 4096,
1922
+ },
1923
+ "qwen/qwen3-vl-30b-a3b-thinking": {
1924
+ id: "qwen/qwen3-vl-30b-a3b-thinking",
1925
+ name: "Qwen: Qwen3 VL 30B A3B Thinking",
1926
+ api: "openai-completions",
1927
+ provider: "openrouter",
1928
+ baseUrl: "https://openrouter.ai/api/v1",
1929
+ reasoning: true,
1930
+ input: ["text", "image"],
1931
+ cost: {
1932
+ input: 0.19999999999999998,
1933
+ output: 1,
1934
+ cacheRead: 0,
1935
+ cacheWrite: 0,
1936
+ },
1937
+ contextWindow: 131072,
1938
+ maxTokens: 32768,
1939
+ },
1940
+ "qwen/qwen3-vl-30b-a3b-instruct": {
1941
+ id: "qwen/qwen3-vl-30b-a3b-instruct",
1942
+ name: "Qwen: Qwen3 VL 30B A3B Instruct",
1943
+ api: "openai-completions",
1944
+ provider: "openrouter",
1945
+ baseUrl: "https://openrouter.ai/api/v1",
1946
+ reasoning: false,
1947
+ input: ["text", "image"],
1948
+ cost: {
1949
+ input: 0.19999999999999998,
1950
+ output: 0.7,
1951
+ cacheRead: 0,
1952
+ cacheWrite: 0,
1953
+ },
1954
+ contextWindow: 131072,
1955
+ maxTokens: 32768,
1956
+ },
1651
1957
  "z-ai/glm-4.6": {
1652
1958
  id: "z-ai/glm-4.6",
1653
1959
  name: "Z.AI: GLM 4.6",
@@ -1691,13 +1997,13 @@ export const MODELS = {
1691
1997
  reasoning: true,
1692
1998
  input: ["text", "image"],
1693
1999
  cost: {
1694
- input: 0.5,
1695
- output: 3.5,
2000
+ input: 0.3,
2001
+ output: 1.2,
1696
2002
  cacheRead: 0,
1697
2003
  cacheWrite: 0,
1698
2004
  },
1699
- contextWindow: 65536,
1700
- maxTokens: 65536,
2005
+ contextWindow: 262144,
2006
+ maxTokens: 262144,
1701
2007
  },
1702
2008
  "qwen/qwen3-vl-235b-a22b-instruct": {
1703
2009
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -1709,12 +2015,12 @@ export const MODELS = {
1709
2015
  input: ["text", "image"],
1710
2016
  cost: {
1711
2017
  input: 0.3,
1712
- output: 1.5,
2018
+ output: 1.2,
1713
2019
  cacheRead: 0,
1714
2020
  cacheWrite: 0,
1715
2021
  },
1716
- contextWindow: 131072,
1717
- maxTokens: 32768,
2022
+ contextWindow: 262144,
2023
+ maxTokens: 262144,
1718
2024
  },
1719
2025
  "qwen/qwen3-max": {
1720
2026
  id: "qwen/qwen3-max",
@@ -1773,7 +2079,7 @@ export const MODELS = {
1773
2079
  api: "openai-completions",
1774
2080
  provider: "openrouter",
1775
2081
  baseUrl: "https://openrouter.ai/api/v1",
1776
- reasoning: false,
2082
+ reasoning: true,
1777
2083
  input: ["text"],
1778
2084
  cost: {
1779
2085
  input: 0,
@@ -1827,13 +2133,13 @@ export const MODELS = {
1827
2133
  reasoning: true,
1828
2134
  input: ["text"],
1829
2135
  cost: {
1830
- input: 0.09999999999999999,
1831
- output: 0.7999999999999999,
2136
+ input: 0.14,
2137
+ output: 1.2,
1832
2138
  cacheRead: 0,
1833
2139
  cacheWrite: 0,
1834
2140
  },
1835
2141
  contextWindow: 262144,
1836
- maxTokens: 262144,
2142
+ maxTokens: 4096,
1837
2143
  },
1838
2144
  "qwen/qwen3-next-80b-a3b-instruct": {
1839
2145
  id: "qwen/qwen3-next-80b-a3b-instruct",
@@ -1954,6 +2260,23 @@ export const MODELS = {
1954
2260
  contextWindow: 262144,
1955
2261
  maxTokens: 262144,
1956
2262
  },
2263
+ "deepcogito/cogito-v2-preview-llama-70b": {
2264
+ id: "deepcogito/cogito-v2-preview-llama-70b",
2265
+ name: "Deep Cogito: Cogito V2 Preview Llama 70B",
2266
+ api: "openai-completions",
2267
+ provider: "openrouter",
2268
+ baseUrl: "https://openrouter.ai/api/v1",
2269
+ reasoning: true,
2270
+ input: ["text"],
2271
+ cost: {
2272
+ input: 0.88,
2273
+ output: 0.88,
2274
+ cacheRead: 0,
2275
+ cacheWrite: 0,
2276
+ },
2277
+ contextWindow: 32768,
2278
+ maxTokens: 4096,
2279
+ },
1957
2280
  "deepcogito/cogito-v2-preview-llama-109b-moe": {
1958
2281
  id: "deepcogito/cogito-v2-preview-llama-109b-moe",
1959
2282
  name: "Cogito V2 Preview Llama 109B",
@@ -2039,23 +2362,6 @@ export const MODELS = {
2039
2362
  contextWindow: 131072,
2040
2363
  maxTokens: 131072,
2041
2364
  },
2042
- "deepseek/deepseek-chat-v3.1:free": {
2043
- id: "deepseek/deepseek-chat-v3.1:free",
2044
- name: "DeepSeek: DeepSeek V3.1 (free)",
2045
- api: "openai-completions",
2046
- provider: "openrouter",
2047
- baseUrl: "https://openrouter.ai/api/v1",
2048
- reasoning: true,
2049
- input: ["text"],
2050
- cost: {
2051
- input: 0,
2052
- output: 0,
2053
- cacheRead: 0,
2054
- cacheWrite: 0,
2055
- },
2056
- contextWindow: 163800,
2057
- maxTokens: 4096,
2058
- },
2059
2365
  "deepseek/deepseek-chat-v3.1": {
2060
2366
  id: "deepseek/deepseek-chat-v3.1",
2061
2367
  name: "DeepSeek: DeepSeek V3.1",
@@ -2065,13 +2371,13 @@ export const MODELS = {
2065
2371
  reasoning: true,
2066
2372
  input: ["text"],
2067
2373
  cost: {
2068
- input: 0.19999999999999998,
2069
- output: 0.7999999999999999,
2374
+ input: 0.27,
2375
+ output: 1,
2070
2376
  cacheRead: 0,
2071
2377
  cacheWrite: 0,
2072
2378
  },
2073
- contextWindow: 163840,
2074
- maxTokens: 163840,
2379
+ contextWindow: 131072,
2380
+ maxTokens: 32768,
2075
2381
  },
2076
2382
  "mistralai/mistral-medium-3.1": {
2077
2383
  id: "mistralai/mistral-medium-3.1",
@@ -2090,6 +2396,40 @@ export const MODELS = {
2090
2396
  contextWindow: 131072,
2091
2397
  maxTokens: 4096,
2092
2398
  },
2399
+ "baidu/ernie-4.5-21b-a3b": {
2400
+ id: "baidu/ernie-4.5-21b-a3b",
2401
+ name: "Baidu: ERNIE 4.5 21B A3B",
2402
+ api: "openai-completions",
2403
+ provider: "openrouter",
2404
+ baseUrl: "https://openrouter.ai/api/v1",
2405
+ reasoning: false,
2406
+ input: ["text"],
2407
+ cost: {
2408
+ input: 0.07,
2409
+ output: 0.28,
2410
+ cacheRead: 0,
2411
+ cacheWrite: 0,
2412
+ },
2413
+ contextWindow: 120000,
2414
+ maxTokens: 8000,
2415
+ },
2416
+ "baidu/ernie-4.5-vl-28b-a3b": {
2417
+ id: "baidu/ernie-4.5-vl-28b-a3b",
2418
+ name: "Baidu: ERNIE 4.5 VL 28B A3B",
2419
+ api: "openai-completions",
2420
+ provider: "openrouter",
2421
+ baseUrl: "https://openrouter.ai/api/v1",
2422
+ reasoning: true,
2423
+ input: ["text", "image"],
2424
+ cost: {
2425
+ input: 0.14,
2426
+ output: 0.56,
2427
+ cacheRead: 0,
2428
+ cacheWrite: 0,
2429
+ },
2430
+ contextWindow: 30000,
2431
+ maxTokens: 8000,
2432
+ },
2093
2433
  "z-ai/glm-4.5v": {
2094
2434
  id: "z-ai/glm-4.5v",
2095
2435
  name: "Z.AI: GLM 4.5V",
@@ -2101,7 +2441,7 @@ export const MODELS = {
2101
2441
  cost: {
2102
2442
  input: 0.6,
2103
2443
  output: 1.7999999999999998,
2104
- cacheRead: 0,
2444
+ cacheRead: 0.11,
2105
2445
  cacheWrite: 0,
2106
2446
  },
2107
2447
  contextWindow: 65536,
@@ -2202,7 +2542,7 @@ export const MODELS = {
2202
2542
  input: ["text"],
2203
2543
  cost: {
2204
2544
  input: 0.35,
2205
- output: 1.55,
2545
+ output: 1.5,
2206
2546
  cacheRead: 0,
2207
2547
  cacheWrite: 0,
2208
2548
  },
@@ -2235,13 +2575,13 @@ export const MODELS = {
2235
2575
  reasoning: true,
2236
2576
  input: ["text"],
2237
2577
  cost: {
2238
- input: 0.14,
2239
- output: 0.86,
2578
+ input: 0.13,
2579
+ output: 0.85,
2240
2580
  cacheRead: 0,
2241
2581
  cacheWrite: 0,
2242
2582
  },
2243
2583
  contextWindow: 131072,
2244
- maxTokens: 131072,
2584
+ maxTokens: 98304,
2245
2585
  },
2246
2586
  "qwen/qwen3-235b-a22b-thinking-2507": {
2247
2587
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -2291,8 +2631,8 @@ export const MODELS = {
2291
2631
  cacheRead: 0,
2292
2632
  cacheWrite: 0,
2293
2633
  },
2294
- contextWindow: 262144,
2295
- maxTokens: 4096,
2634
+ contextWindow: 262000,
2635
+ maxTokens: 262000,
2296
2636
  },
2297
2637
  "qwen/qwen3-coder": {
2298
2638
  id: "qwen/qwen3-coder",
@@ -2300,7 +2640,7 @@ export const MODELS = {
2300
2640
  api: "openai-completions",
2301
2641
  provider: "openrouter",
2302
2642
  baseUrl: "https://openrouter.ai/api/v1",
2303
- reasoning: false,
2643
+ reasoning: true,
2304
2644
  input: ["text"],
2305
2645
  cost: {
2306
2646
  input: 0.22,
@@ -2317,7 +2657,7 @@ export const MODELS = {
2317
2657
  api: "openai-completions",
2318
2658
  provider: "openrouter",
2319
2659
  baseUrl: "https://openrouter.ai/api/v1",
2320
- reasoning: false,
2660
+ reasoning: true,
2321
2661
  input: ["text"],
2322
2662
  cost: {
2323
2663
  input: 0.08,
@@ -2379,6 +2719,23 @@ export const MODELS = {
2379
2719
  contextWindow: 128000,
2380
2720
  maxTokens: 4096,
2381
2721
  },
2722
+ "tngtech/deepseek-r1t2-chimera": {
2723
+ id: "tngtech/deepseek-r1t2-chimera",
2724
+ name: "TNG: DeepSeek R1T2 Chimera",
2725
+ api: "openai-completions",
2726
+ provider: "openrouter",
2727
+ baseUrl: "https://openrouter.ai/api/v1",
2728
+ reasoning: true,
2729
+ input: ["text"],
2730
+ cost: {
2731
+ input: 0.3,
2732
+ output: 1.2,
2733
+ cacheRead: 0,
2734
+ cacheWrite: 0,
2735
+ },
2736
+ contextWindow: 163840,
2737
+ maxTokens: 163840,
2738
+ },
2382
2739
  "inception/mercury": {
2383
2740
  id: "inception/mercury",
2384
2741
  name: "Inception: Mercury",
@@ -2439,8 +2796,8 @@ export const MODELS = {
2439
2796
  reasoning: true,
2440
2797
  input: ["text"],
2441
2798
  cost: {
2442
- input: 0.3,
2443
- output: 1.6500000000000001,
2799
+ input: 0.39999999999999997,
2800
+ output: 2.2,
2444
2801
  cacheRead: 0,
2445
2802
  cacheWrite: 0,
2446
2803
  },
@@ -2566,6 +2923,23 @@ export const MODELS = {
2566
2923
  contextWindow: 128000,
2567
2924
  maxTokens: 4028,
2568
2925
  },
2926
+ "nousresearch/deephermes-3-mistral-24b-preview": {
2927
+ id: "nousresearch/deephermes-3-mistral-24b-preview",
2928
+ name: "Nous: DeepHermes 3 Mistral 24B Preview",
2929
+ api: "openai-completions",
2930
+ provider: "openrouter",
2931
+ baseUrl: "https://openrouter.ai/api/v1",
2932
+ reasoning: true,
2933
+ input: ["text"],
2934
+ cost: {
2935
+ input: 0.15,
2936
+ output: 0.59,
2937
+ cacheRead: 0,
2938
+ cacheWrite: 0,
2939
+ },
2940
+ contextWindow: 32768,
2941
+ maxTokens: 32768,
2942
+ },
2569
2943
  "mistralai/mistral-medium-3": {
2570
2944
  id: "mistralai/mistral-medium-3",
2571
2945
  name: "Mistral: Mistral Medium 3",
@@ -2699,7 +3073,7 @@ export const MODELS = {
2699
3073
  cacheRead: 0,
2700
3074
  cacheWrite: 0,
2701
3075
  },
2702
- contextWindow: 131072,
3076
+ contextWindow: 40960,
2703
3077
  maxTokens: 4096,
2704
3078
  },
2705
3079
  "qwen/qwen3-235b-a22b": {
@@ -2810,7 +3184,7 @@ export const MODELS = {
2810
3184
  api: "openai-completions",
2811
3185
  provider: "openrouter",
2812
3186
  baseUrl: "https://openrouter.ai/api/v1",
2813
- reasoning: false,
3187
+ reasoning: true,
2814
3188
  input: ["text"],
2815
3189
  cost: {
2816
3190
  input: 0.24,
@@ -2855,39 +3229,39 @@ export const MODELS = {
2855
3229
  contextWindow: 128000,
2856
3230
  maxTokens: 4096,
2857
3231
  },
2858
- "microsoft/phi-4-multimodal-instruct": {
2859
- id: "microsoft/phi-4-multimodal-instruct",
2860
- name: "Microsoft: Phi 4 Multimodal Instruct",
3232
+ "qwen/qwq-32b": {
3233
+ id: "qwen/qwq-32b",
3234
+ name: "Qwen: QwQ 32B",
2861
3235
  api: "openai-completions",
2862
3236
  provider: "openrouter",
2863
3237
  baseUrl: "https://openrouter.ai/api/v1",
2864
- reasoning: false,
2865
- input: ["text", "image"],
3238
+ reasoning: true,
3239
+ input: ["text"],
2866
3240
  cost: {
2867
- input: 0.049999999999999996,
2868
- output: 0.09999999999999999,
3241
+ input: 0.15,
3242
+ output: 0.39999999999999997,
2869
3243
  cacheRead: 0,
2870
3244
  cacheWrite: 0,
2871
3245
  },
2872
- contextWindow: 131072,
3246
+ contextWindow: 32768,
2873
3247
  maxTokens: 4096,
2874
3248
  },
2875
- "qwen/qwq-32b": {
2876
- id: "qwen/qwq-32b",
2877
- name: "Qwen: QwQ 32B",
3249
+ "nousresearch/deephermes-3-llama-3-8b-preview": {
3250
+ id: "nousresearch/deephermes-3-llama-3-8b-preview",
3251
+ name: "Nous: DeepHermes 3 Llama 3 8B Preview",
2878
3252
  api: "openai-completions",
2879
3253
  provider: "openrouter",
2880
3254
  baseUrl: "https://openrouter.ai/api/v1",
2881
- reasoning: true,
3255
+ reasoning: false,
2882
3256
  input: ["text"],
2883
3257
  cost: {
2884
- input: 0.15,
2885
- output: 0.39999999999999997,
3258
+ input: 0.03,
3259
+ output: 0.11,
2886
3260
  cacheRead: 0,
2887
3261
  cacheWrite: 0,
2888
3262
  },
2889
- contextWindow: 32768,
2890
- maxTokens: 4096,
3263
+ contextWindow: 131072,
3264
+ maxTokens: 131072,
2891
3265
  },
2892
3266
  "mistralai/mistral-saba": {
2893
3267
  id: "mistralai/mistral-saba",
@@ -2906,6 +3280,23 @@ export const MODELS = {
2906
3280
  contextWindow: 32768,
2907
3281
  maxTokens: 4096,
2908
3282
  },
3283
+ "qwen/qwen-vl-max": {
3284
+ id: "qwen/qwen-vl-max",
3285
+ name: "Qwen: Qwen VL Max",
3286
+ api: "openai-completions",
3287
+ provider: "openrouter",
3288
+ baseUrl: "https://openrouter.ai/api/v1",
3289
+ reasoning: false,
3290
+ input: ["text", "image"],
3291
+ cost: {
3292
+ input: 0.7999999999999999,
3293
+ output: 3.1999999999999997,
3294
+ cacheRead: 0,
3295
+ cacheWrite: 0,
3296
+ },
3297
+ contextWindow: 131072,
3298
+ maxTokens: 8192,
3299
+ },
2909
3300
  "qwen/qwen-turbo": {
2910
3301
  id: "qwen/qwen-turbo",
2911
3302
  name: "Qwen: Qwen-Turbo",
@@ -3056,8 +3447,8 @@ export const MODELS = {
3056
3447
  cacheRead: 0,
3057
3448
  cacheWrite: 0,
3058
3449
  },
3059
- contextWindow: 65536,
3060
- maxTokens: 4096,
3450
+ contextWindow: 131072,
3451
+ maxTokens: 2048,
3061
3452
  },
3062
3453
  "meta-llama/llama-3.3-70b-instruct": {
3063
3454
  id: "meta-llama/llama-3.3-70b-instruct",
@@ -3068,13 +3459,13 @@ export const MODELS = {
3068
3459
  reasoning: false,
3069
3460
  input: ["text"],
3070
3461
  cost: {
3071
- input: 0.04,
3072
- output: 0.12,
3462
+ input: 0.13,
3463
+ output: 0.38,
3073
3464
  cacheRead: 0,
3074
3465
  cacheWrite: 0,
3075
3466
  },
3076
3467
  contextWindow: 131072,
3077
- maxTokens: 131072,
3468
+ maxTokens: 16384,
3078
3469
  },
3079
3470
  "amazon/nova-lite-v1": {
3080
3471
  id: "amazon/nova-lite-v1",
@@ -3212,6 +3603,23 @@ export const MODELS = {
3212
3603
  contextWindow: 128000,
3213
3604
  maxTokens: 4096,
3214
3605
  },
3606
+ "qwen/qwen-2.5-7b-instruct": {
3607
+ id: "qwen/qwen-2.5-7b-instruct",
3608
+ name: "Qwen: Qwen2.5 7B Instruct",
3609
+ api: "openai-completions",
3610
+ provider: "openrouter",
3611
+ baseUrl: "https://openrouter.ai/api/v1",
3612
+ reasoning: false,
3613
+ input: ["text"],
3614
+ cost: {
3615
+ input: 0.04,
3616
+ output: 0.09999999999999999,
3617
+ cacheRead: 0,
3618
+ cacheWrite: 0,
3619
+ },
3620
+ contextWindow: 32768,
3621
+ maxTokens: 16384,
3622
+ },
3215
3623
  "nvidia/llama-3.1-nemotron-70b-instruct": {
3216
3624
  id: "nvidia/llama-3.1-nemotron-70b-instruct",
3217
3625
  name: "NVIDIA: Llama 3.1 Nemotron 70B Instruct",
@@ -3331,6 +3739,23 @@ export const MODELS = {
3331
3739
  contextWindow: 128000,
3332
3740
  maxTokens: 4000,
3333
3741
  },
3742
+ "sao10k/l3.1-euryale-70b": {
3743
+ id: "sao10k/l3.1-euryale-70b",
3744
+ name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
3745
+ api: "openai-completions",
3746
+ provider: "openrouter",
3747
+ baseUrl: "https://openrouter.ai/api/v1",
3748
+ reasoning: false,
3749
+ input: ["text"],
3750
+ cost: {
3751
+ input: 0.65,
3752
+ output: 0.75,
3753
+ cacheRead: 0,
3754
+ cacheWrite: 0,
3755
+ },
3756
+ contextWindow: 32768,
3757
+ maxTokens: 4096,
3758
+ },
3334
3759
  "microsoft/phi-3.5-mini-128k-instruct": {
3335
3760
  id: "microsoft/phi-3.5-mini-128k-instruct",
3336
3761
  name: "Microsoft: Phi-3.5 Mini 128K Instruct",
@@ -3362,41 +3787,41 @@ export const MODELS = {
3362
3787
  cacheRead: 0,
3363
3788
  cacheWrite: 0,
3364
3789
  },
3365
- contextWindow: 65000,
3790
+ contextWindow: 65536,
3366
3791
  maxTokens: 4096,
3367
3792
  },
3368
- "meta-llama/llama-3.1-405b-instruct": {
3369
- id: "meta-llama/llama-3.1-405b-instruct",
3370
- name: "Meta: Llama 3.1 405B Instruct",
3793
+ "meta-llama/llama-3.1-8b-instruct": {
3794
+ id: "meta-llama/llama-3.1-8b-instruct",
3795
+ name: "Meta: Llama 3.1 8B Instruct",
3371
3796
  api: "openai-completions",
3372
3797
  provider: "openrouter",
3373
3798
  baseUrl: "https://openrouter.ai/api/v1",
3374
3799
  reasoning: false,
3375
3800
  input: ["text"],
3376
3801
  cost: {
3377
- input: 0.7999999999999999,
3378
- output: 0.7999999999999999,
3802
+ input: 0.02,
3803
+ output: 0.03,
3379
3804
  cacheRead: 0,
3380
3805
  cacheWrite: 0,
3381
3806
  },
3382
- contextWindow: 32768,
3807
+ contextWindow: 16384,
3383
3808
  maxTokens: 16384,
3384
3809
  },
3385
- "meta-llama/llama-3.1-8b-instruct": {
3386
- id: "meta-llama/llama-3.1-8b-instruct",
3387
- name: "Meta: Llama 3.1 8B Instruct",
3810
+ "meta-llama/llama-3.1-405b-instruct": {
3811
+ id: "meta-llama/llama-3.1-405b-instruct",
3812
+ name: "Meta: Llama 3.1 405B Instruct",
3388
3813
  api: "openai-completions",
3389
3814
  provider: "openrouter",
3390
3815
  baseUrl: "https://openrouter.ai/api/v1",
3391
3816
  reasoning: false,
3392
3817
  input: ["text"],
3393
3818
  cost: {
3394
- input: 0.02,
3395
- output: 0.03,
3819
+ input: 0.7999999999999999,
3820
+ output: 0.7999999999999999,
3396
3821
  cacheRead: 0,
3397
3822
  cacheWrite: 0,
3398
3823
  },
3399
- contextWindow: 16384,
3824
+ contextWindow: 32768,
3400
3825
  maxTokens: 16384,
3401
3826
  },
3402
3827
  "meta-llama/llama-3.1-70b-instruct": {
@@ -3433,6 +3858,23 @@ export const MODELS = {
3433
3858
  contextWindow: 131072,
3434
3859
  maxTokens: 16384,
3435
3860
  },
3861
+ "sao10k/l3-euryale-70b": {
3862
+ id: "sao10k/l3-euryale-70b",
3863
+ name: "Sao10k: Llama 3 Euryale 70B v2.1",
3864
+ api: "openai-completions",
3865
+ provider: "openrouter",
3866
+ baseUrl: "https://openrouter.ai/api/v1",
3867
+ reasoning: false,
3868
+ input: ["text"],
3869
+ cost: {
3870
+ input: 1.48,
3871
+ output: 1.48,
3872
+ cacheRead: 0,
3873
+ cacheWrite: 0,
3874
+ },
3875
+ contextWindow: 8192,
3876
+ maxTokens: 8192,
3877
+ },
3436
3878
  "mistralai/mistral-7b-instruct:free": {
3437
3879
  id: "mistralai/mistral-7b-instruct:free",
3438
3880
  name: "Mistral: Mistral 7B Instruct (free)",
@@ -3561,8 +4003,8 @@ export const MODELS = {
3561
4003
  reasoning: false,
3562
4004
  input: ["text"],
3563
4005
  cost: {
3564
- input: 0.8999999999999999,
3565
- output: 0.8999999999999999,
4006
+ input: 2,
4007
+ output: 6,
3566
4008
  cacheRead: 0,
3567
4009
  cacheWrite: 0,
3568
4010
  },