@hyperspaceng/neural-ai 0.64.1 → 0.65.3

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.
@@ -1551,23 +1551,6 @@ export const MODELS = {
1551
1551
  contextWindow: 200000,
1552
1552
  maxTokens: 64000,
1553
1553
  },
1554
- "claude-3-7-sonnet-latest": {
1555
- id: "claude-3-7-sonnet-latest",
1556
- name: "Claude Sonnet 3.7 (latest)",
1557
- api: "anthropic-messages",
1558
- provider: "anthropic",
1559
- baseUrl: "https://api.anthropic.com",
1560
- reasoning: true,
1561
- input: ["text", "image"],
1562
- cost: {
1563
- input: 3,
1564
- output: 15,
1565
- cacheRead: 0.3,
1566
- cacheWrite: 3.75,
1567
- },
1568
- contextWindow: 200000,
1569
- maxTokens: 64000,
1570
- },
1571
1554
  "claude-3-haiku-20240307": {
1572
1555
  id: "claude-3-haiku-20240307",
1573
1556
  name: "Claude Haiku 3",
@@ -3032,7 +3015,7 @@ export const MODELS = {
3032
3015
  },
3033
3016
  "gpt-5.4-mini": {
3034
3017
  id: "gpt-5.4-mini",
3035
- name: "GPT-5.4 mini",
3018
+ name: "GPT-5.4 Mini",
3036
3019
  api: "openai-responses",
3037
3020
  provider: "github-copilot",
3038
3021
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3494,6 +3477,40 @@ export const MODELS = {
3494
3477
  contextWindow: 131072,
3495
3478
  maxTokens: 8192,
3496
3479
  },
3480
+ "gemma-4-26b-it": {
3481
+ id: "gemma-4-26b-it",
3482
+ name: "Gemma 4 26B",
3483
+ api: "google-generative-ai",
3484
+ provider: "google",
3485
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
3486
+ reasoning: true,
3487
+ input: ["text", "image"],
3488
+ cost: {
3489
+ input: 0,
3490
+ output: 0,
3491
+ cacheRead: 0,
3492
+ cacheWrite: 0,
3493
+ },
3494
+ contextWindow: 256000,
3495
+ maxTokens: 8192,
3496
+ },
3497
+ "gemma-4-31b-it": {
3498
+ id: "gemma-4-31b-it",
3499
+ name: "Gemma 4 31B",
3500
+ api: "google-generative-ai",
3501
+ provider: "google",
3502
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
3503
+ reasoning: true,
3504
+ input: ["text", "image"],
3505
+ cost: {
3506
+ input: 0,
3507
+ output: 0,
3508
+ cacheRead: 0,
3509
+ cacheWrite: 0,
3510
+ },
3511
+ contextWindow: 256000,
3512
+ maxTokens: 8192,
3513
+ },
3497
3514
  },
3498
3515
  "google-antigravity": {
3499
3516
  "claude-opus-4-5-thinking": {
@@ -6477,40 +6494,6 @@ export const MODELS = {
6477
6494
  contextWindow: 262144,
6478
6495
  maxTokens: 65536,
6479
6496
  },
6480
- "mimo-v2-omni-free": {
6481
- id: "mimo-v2-omni-free",
6482
- name: "MiMo V2 Omni Free",
6483
- api: "openai-completions",
6484
- provider: "opencode",
6485
- baseUrl: "https://opencode.ai/zen/v1",
6486
- reasoning: true,
6487
- input: ["text", "image"],
6488
- cost: {
6489
- input: 0,
6490
- output: 0,
6491
- cacheRead: 0,
6492
- cacheWrite: 0,
6493
- },
6494
- contextWindow: 262144,
6495
- maxTokens: 64000,
6496
- },
6497
- "mimo-v2-pro-free": {
6498
- id: "mimo-v2-pro-free",
6499
- name: "MiMo V2 Pro Free",
6500
- api: "openai-completions",
6501
- provider: "opencode",
6502
- baseUrl: "https://opencode.ai/zen/v1",
6503
- reasoning: true,
6504
- input: ["text"],
6505
- cost: {
6506
- input: 0,
6507
- output: 0,
6508
- cacheRead: 0,
6509
- cacheWrite: 0,
6510
- },
6511
- contextWindow: 1048576,
6512
- maxTokens: 64000,
6513
- },
6514
6497
  "minimax-m2.5": {
6515
6498
  id: "minimax-m2.5",
6516
6499
  name: "MiniMax M2.5",
@@ -6559,7 +6542,7 @@ export const MODELS = {
6559
6542
  cacheRead: 0,
6560
6543
  cacheWrite: 0,
6561
6544
  },
6562
- contextWindow: 1000000,
6545
+ contextWindow: 204800,
6563
6546
  maxTokens: 128000,
6564
6547
  },
6565
6548
  "qwen3.6-plus-free": {
@@ -6615,12 +6598,46 @@ export const MODELS = {
6615
6598
  contextWindow: 262144,
6616
6599
  maxTokens: 65536,
6617
6600
  },
6601
+ "mimo-v2-omni": {
6602
+ id: "mimo-v2-omni",
6603
+ name: "MiMo V2 Omni",
6604
+ api: "openai-completions",
6605
+ provider: "opencode-go",
6606
+ baseUrl: "https://opencode.ai/zen/go/v1",
6607
+ reasoning: true,
6608
+ input: ["text", "image"],
6609
+ cost: {
6610
+ input: 0.4,
6611
+ output: 2,
6612
+ cacheRead: 0.08,
6613
+ cacheWrite: 0,
6614
+ },
6615
+ contextWindow: 262144,
6616
+ maxTokens: 64000,
6617
+ },
6618
+ "mimo-v2-pro": {
6619
+ id: "mimo-v2-pro",
6620
+ name: "MiMo V2 Pro",
6621
+ api: "openai-completions",
6622
+ provider: "opencode-go",
6623
+ baseUrl: "https://opencode.ai/zen/go/v1",
6624
+ reasoning: true,
6625
+ input: ["text"],
6626
+ cost: {
6627
+ input: 1,
6628
+ output: 3,
6629
+ cacheRead: 0.2,
6630
+ cacheWrite: 0,
6631
+ },
6632
+ contextWindow: 1048576,
6633
+ maxTokens: 64000,
6634
+ },
6618
6635
  "minimax-m2.5": {
6619
6636
  id: "minimax-m2.5",
6620
6637
  name: "MiniMax M2.5",
6621
- api: "anthropic-messages",
6638
+ api: "openai-completions",
6622
6639
  provider: "opencode-go",
6623
- baseUrl: "https://opencode.ai/zen/go",
6640
+ baseUrl: "https://opencode.ai/zen/go/v1",
6624
6641
  reasoning: true,
6625
6642
  input: ["text"],
6626
6643
  cost: {
@@ -6821,23 +6838,6 @@ export const MODELS = {
6821
6838
  contextWindow: 200000,
6822
6839
  maxTokens: 8192,
6823
6840
  },
6824
- "anthropic/claude-3.5-sonnet": {
6825
- id: "anthropic/claude-3.5-sonnet",
6826
- name: "Anthropic: Claude 3.5 Sonnet",
6827
- api: "openai-completions",
6828
- provider: "openrouter",
6829
- baseUrl: "https://openrouter.ai/api/v1",
6830
- reasoning: false,
6831
- input: ["text", "image"],
6832
- cost: {
6833
- input: 6,
6834
- output: 30,
6835
- cacheRead: 0.6,
6836
- cacheWrite: 7.5,
6837
- },
6838
- contextWindow: 200000,
6839
- maxTokens: 8192,
6840
- },
6841
6841
  "anthropic/claude-3.7-sonnet": {
6842
6842
  id: "anthropic/claude-3.7-sonnet",
6843
6843
  name: "Anthropic: Claude 3.7 Sonnet",
@@ -7025,6 +7025,23 @@ export const MODELS = {
7025
7025
  contextWindow: 131000,
7026
7026
  maxTokens: 4096,
7027
7027
  },
7028
+ "arcee-ai/trinity-large-thinking": {
7029
+ id: "arcee-ai/trinity-large-thinking",
7030
+ name: "Arcee AI: Trinity Large Thinking",
7031
+ api: "openai-completions",
7032
+ provider: "openrouter",
7033
+ baseUrl: "https://openrouter.ai/api/v1",
7034
+ reasoning: true,
7035
+ input: ["text"],
7036
+ cost: {
7037
+ input: 0.22,
7038
+ output: 0.85,
7039
+ cacheRead: 0,
7040
+ cacheWrite: 0,
7041
+ },
7042
+ contextWindow: 262144,
7043
+ maxTokens: 262144,
7044
+ },
7028
7045
  "arcee-ai/trinity-mini": {
7029
7046
  id: "arcee-ai/trinity-mini",
7030
7047
  name: "Arcee AI: Trinity Mini",
@@ -7465,7 +7482,7 @@ export const MODELS = {
7465
7482
  cacheWrite: 0.08333333333333334,
7466
7483
  },
7467
7484
  contextWindow: 1048576,
7468
- maxTokens: 65536,
7485
+ maxTokens: 65535,
7469
7486
  },
7470
7487
  "google/gemini-2.5-pro": {
7471
7488
  id: "google/gemini-2.5-pro",
@@ -7586,6 +7603,40 @@ export const MODELS = {
7586
7603
  contextWindow: 1048576,
7587
7604
  maxTokens: 65536,
7588
7605
  },
7606
+ "google/gemma-4-26b-a4b-it": {
7607
+ id: "google/gemma-4-26b-a4b-it",
7608
+ name: "Google: Gemma 4 26B A4B ",
7609
+ api: "openai-completions",
7610
+ provider: "openrouter",
7611
+ baseUrl: "https://openrouter.ai/api/v1",
7612
+ reasoning: true,
7613
+ input: ["text", "image"],
7614
+ cost: {
7615
+ input: 0.13,
7616
+ output: 0.39999999999999997,
7617
+ cacheRead: 0,
7618
+ cacheWrite: 0,
7619
+ },
7620
+ contextWindow: 262144,
7621
+ maxTokens: 262144,
7622
+ },
7623
+ "google/gemma-4-31b-it": {
7624
+ id: "google/gemma-4-31b-it",
7625
+ name: "Google: Gemma 4 31B",
7626
+ api: "openai-completions",
7627
+ provider: "openrouter",
7628
+ baseUrl: "https://openrouter.ai/api/v1",
7629
+ reasoning: true,
7630
+ input: ["text", "image"],
7631
+ cost: {
7632
+ input: 0.14,
7633
+ output: 0.39999999999999997,
7634
+ cacheRead: 0,
7635
+ cacheWrite: 0,
7636
+ },
7637
+ contextWindow: 262144,
7638
+ maxTokens: 131072,
7639
+ },
7589
7640
  "inception/mercury": {
7590
7641
  id: "inception/mercury",
7591
7642
  name: "Inception: Mercury",
@@ -7850,9 +7901,9 @@ export const MODELS = {
7850
7901
  reasoning: true,
7851
7902
  input: ["text"],
7852
7903
  cost: {
7853
- input: 0.19,
7854
- output: 1.15,
7855
- cacheRead: 0.095,
7904
+ input: 0.118,
7905
+ output: 0.9900000000000001,
7906
+ cacheRead: 0.059,
7856
7907
  cacheWrite: 0,
7857
7908
  },
7858
7909
  contextWindow: 196608,
@@ -8147,23 +8198,6 @@ export const MODELS = {
8147
8198
  contextWindow: 32768,
8148
8199
  maxTokens: 4096,
8149
8200
  },
8150
- "mistralai/mistral-small-24b-instruct-2501": {
8151
- id: "mistralai/mistral-small-24b-instruct-2501",
8152
- name: "Mistral: Mistral Small 3",
8153
- api: "openai-completions",
8154
- provider: "openrouter",
8155
- baseUrl: "https://openrouter.ai/api/v1",
8156
- reasoning: false,
8157
- input: ["text"],
8158
- cost: {
8159
- input: 0.049999999999999996,
8160
- output: 0.08,
8161
- cacheRead: 0,
8162
- cacheWrite: 0,
8163
- },
8164
- contextWindow: 32768,
8165
- maxTokens: 16384,
8166
- },
8167
8201
  "mistralai/mistral-small-2603": {
8168
8202
  id: "mistralai/mistral-small-2603",
8169
8203
  name: "Mistral: Mistral Small 4",
@@ -9218,6 +9252,40 @@ export const MODELS = {
9218
9252
  contextWindow: 1050000,
9219
9253
  maxTokens: 128000,
9220
9254
  },
9255
+ "openai/gpt-audio": {
9256
+ id: "openai/gpt-audio",
9257
+ name: "OpenAI: GPT Audio",
9258
+ api: "openai-completions",
9259
+ provider: "openrouter",
9260
+ baseUrl: "https://openrouter.ai/api/v1",
9261
+ reasoning: false,
9262
+ input: ["text"],
9263
+ cost: {
9264
+ input: 2.5,
9265
+ output: 10,
9266
+ cacheRead: 0,
9267
+ cacheWrite: 0,
9268
+ },
9269
+ contextWindow: 128000,
9270
+ maxTokens: 16384,
9271
+ },
9272
+ "openai/gpt-audio-mini": {
9273
+ id: "openai/gpt-audio-mini",
9274
+ name: "OpenAI: GPT Audio Mini",
9275
+ api: "openai-completions",
9276
+ provider: "openrouter",
9277
+ baseUrl: "https://openrouter.ai/api/v1",
9278
+ reasoning: false,
9279
+ input: ["text"],
9280
+ cost: {
9281
+ input: 0.6,
9282
+ output: 2.4,
9283
+ cacheRead: 0,
9284
+ cacheWrite: 0,
9285
+ },
9286
+ contextWindow: 128000,
9287
+ maxTokens: 16384,
9288
+ },
9221
9289
  "openai/gpt-oss-120b": {
9222
9290
  id: "openai/gpt-oss-120b",
9223
9291
  name: "OpenAI: gpt-oss-120b",
@@ -10185,7 +10253,7 @@ export const MODELS = {
10185
10253
  cacheWrite: 0,
10186
10254
  },
10187
10255
  contextWindow: 256000,
10188
- maxTokens: 65536,
10256
+ maxTokens: 32768,
10189
10257
  },
10190
10258
  "qwen/qwen3.5-flash-02-23": {
10191
10259
  id: "qwen/qwen3.5-flash-02-23",
@@ -10221,14 +10289,14 @@ export const MODELS = {
10221
10289
  contextWindow: 1000000,
10222
10290
  maxTokens: 65536,
10223
10291
  },
10224
- "qwen/qwen3.6-plus-preview:free": {
10225
- id: "qwen/qwen3.6-plus-preview:free",
10226
- name: "Qwen: Qwen3.6 Plus Preview (free)",
10292
+ "qwen/qwen3.6-plus:free": {
10293
+ id: "qwen/qwen3.6-plus:free",
10294
+ name: "Qwen: Qwen3.6 Plus (free)",
10227
10295
  api: "openai-completions",
10228
10296
  provider: "openrouter",
10229
10297
  baseUrl: "https://openrouter.ai/api/v1",
10230
10298
  reasoning: true,
10231
- input: ["text"],
10299
+ input: ["text", "image"],
10232
10300
  cost: {
10233
10301
  input: 0,
10234
10302
  output: 0,
@@ -10255,8 +10323,8 @@ export const MODELS = {
10255
10323
  contextWindow: 131072,
10256
10324
  maxTokens: 131072,
10257
10325
  },
10258
- "reka/reka-edge": {
10259
- id: "reka/reka-edge",
10326
+ "rekaai/reka-edge": {
10327
+ id: "rekaai/reka-edge",
10260
10328
  name: "Reka Edge",
10261
10329
  api: "openai-completions",
10262
10330
  provider: "openrouter",
@@ -10544,9 +10612,9 @@ export const MODELS = {
10544
10612
  contextWindow: 2000000,
10545
10613
  maxTokens: 30000,
10546
10614
  },
10547
- "x-ai/grok-4.20-beta": {
10548
- id: "x-ai/grok-4.20-beta",
10549
- name: "xAI: Grok 4.20 Beta",
10615
+ "x-ai/grok-4.20": {
10616
+ id: "x-ai/grok-4.20",
10617
+ name: "xAI: Grok 4.20",
10550
10618
  api: "openai-completions",
10551
10619
  provider: "openrouter",
10552
10620
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10816,6 +10884,23 @@ export const MODELS = {
10816
10884
  contextWindow: 202752,
10817
10885
  maxTokens: 131072,
10818
10886
  },
10887
+ "z-ai/glm-5v-turbo": {
10888
+ id: "z-ai/glm-5v-turbo",
10889
+ name: "Z.ai: GLM 5V Turbo",
10890
+ api: "openai-completions",
10891
+ provider: "openrouter",
10892
+ baseUrl: "https://openrouter.ai/api/v1",
10893
+ reasoning: true,
10894
+ input: ["text", "image"],
10895
+ cost: {
10896
+ input: 1.2,
10897
+ output: 4,
10898
+ cacheRead: 0.24,
10899
+ cacheWrite: 0,
10900
+ },
10901
+ contextWindow: 202752,
10902
+ maxTokens: 131072,
10903
+ },
10819
10904
  },
10820
10905
  "vercel-ai-gateway": {
10821
10906
  "alibaba/qwen-3-14b": {
@@ -10837,7 +10922,7 @@ export const MODELS = {
10837
10922
  },
10838
10923
  "alibaba/qwen-3-235b": {
10839
10924
  id: "alibaba/qwen-3-235b",
10840
- name: "Qwen3-235B-A22B",
10925
+ name: "Qwen3 235B A22b Instruct 2507",
10841
10926
  api: "anthropic-messages",
10842
10927
  provider: "vercel-ai-gateway",
10843
10928
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -11031,13 +11116,13 @@ export const MODELS = {
11031
11116
  reasoning: true,
11032
11117
  input: ["text", "image"],
11033
11118
  cost: {
11034
- input: 0.22,
11035
- output: 0.88,
11119
+ input: 0.39999999999999997,
11120
+ output: 4,
11036
11121
  cacheRead: 0,
11037
11122
  cacheWrite: 0,
11038
11123
  },
11039
- contextWindow: 256000,
11040
- maxTokens: 256000,
11124
+ contextWindow: 131072,
11125
+ maxTokens: 32768,
11041
11126
  },
11042
11127
  "alibaba/qwen3.5-flash": {
11043
11128
  id: "alibaba/qwen3.5-flash",
@@ -11073,6 +11158,23 @@ export const MODELS = {
11073
11158
  contextWindow: 1000000,
11074
11159
  maxTokens: 64000,
11075
11160
  },
11161
+ "alibaba/qwen3.6-plus": {
11162
+ id: "alibaba/qwen3.6-plus",
11163
+ name: "Qwen 3.6 Plus",
11164
+ api: "anthropic-messages",
11165
+ provider: "vercel-ai-gateway",
11166
+ baseUrl: "https://ai-gateway.vercel.sh",
11167
+ reasoning: true,
11168
+ input: ["text", "image"],
11169
+ cost: {
11170
+ input: 0.5,
11171
+ output: 3,
11172
+ cacheRead: 0.09999999999999999,
11173
+ cacheWrite: 0.625,
11174
+ },
11175
+ contextWindow: 1000000,
11176
+ maxTokens: 64000,
11177
+ },
11076
11178
  "anthropic/claude-3-haiku": {
11077
11179
  id: "anthropic/claude-3-haiku",
11078
11180
  name: "Claude 3 Haiku",
@@ -11107,40 +11209,6 @@ export const MODELS = {
11107
11209
  contextWindow: 200000,
11108
11210
  maxTokens: 8192,
11109
11211
  },
11110
- "anthropic/claude-3.5-sonnet": {
11111
- id: "anthropic/claude-3.5-sonnet",
11112
- name: "Claude 3.5 Sonnet",
11113
- api: "anthropic-messages",
11114
- provider: "vercel-ai-gateway",
11115
- baseUrl: "https://ai-gateway.vercel.sh",
11116
- reasoning: false,
11117
- input: ["text", "image"],
11118
- cost: {
11119
- input: 3,
11120
- output: 15,
11121
- cacheRead: 0.3,
11122
- cacheWrite: 3.75,
11123
- },
11124
- contextWindow: 200000,
11125
- maxTokens: 8192,
11126
- },
11127
- "anthropic/claude-3.5-sonnet-20240620": {
11128
- id: "anthropic/claude-3.5-sonnet-20240620",
11129
- name: "Claude 3.5 Sonnet (2024-06-20)",
11130
- api: "anthropic-messages",
11131
- provider: "vercel-ai-gateway",
11132
- baseUrl: "https://ai-gateway.vercel.sh",
11133
- reasoning: false,
11134
- input: ["text", "image"],
11135
- cost: {
11136
- input: 3,
11137
- output: 15,
11138
- cacheRead: 0.3,
11139
- cacheWrite: 3.75,
11140
- },
11141
- contextWindow: 200000,
11142
- maxTokens: 8192,
11143
- },
11144
11212
  "anthropic/claude-3.7-sonnet": {
11145
11213
  id: "anthropic/claude-3.7-sonnet",
11146
11214
  name: "Claude 3.7 Sonnet",
@@ -11311,6 +11379,23 @@ export const MODELS = {
11311
11379
  contextWindow: 131000,
11312
11380
  maxTokens: 131000,
11313
11381
  },
11382
+ "arcee-ai/trinity-large-thinking": {
11383
+ id: "arcee-ai/trinity-large-thinking",
11384
+ name: "Trinity Large Thinking",
11385
+ api: "anthropic-messages",
11386
+ provider: "vercel-ai-gateway",
11387
+ baseUrl: "https://ai-gateway.vercel.sh",
11388
+ reasoning: true,
11389
+ input: ["text"],
11390
+ cost: {
11391
+ input: 0.25,
11392
+ output: 0.8999999999999999,
11393
+ cacheRead: 0,
11394
+ cacheWrite: 0,
11395
+ },
11396
+ contextWindow: 262100,
11397
+ maxTokens: 80000,
11398
+ },
11314
11399
  "bytedance/seed-1.6": {
11315
11400
  id: "bytedance/seed-1.6",
11316
11401
  name: "Seed 1.6",
@@ -11600,6 +11685,40 @@ export const MODELS = {
11600
11685
  contextWindow: 1000000,
11601
11686
  maxTokens: 64000,
11602
11687
  },
11688
+ "google/gemma-4-26b-a4b-it": {
11689
+ id: "google/gemma-4-26b-a4b-it",
11690
+ name: "Gemma 4 26B A4B IT",
11691
+ api: "anthropic-messages",
11692
+ provider: "vercel-ai-gateway",
11693
+ baseUrl: "https://ai-gateway.vercel.sh",
11694
+ reasoning: false,
11695
+ input: ["text", "image"],
11696
+ cost: {
11697
+ input: 0.13,
11698
+ output: 0.39999999999999997,
11699
+ cacheRead: 0,
11700
+ cacheWrite: 0,
11701
+ },
11702
+ contextWindow: 262144,
11703
+ maxTokens: 131072,
11704
+ },
11705
+ "google/gemma-4-31b-it": {
11706
+ id: "google/gemma-4-31b-it",
11707
+ name: "Gemma 4 31B IT",
11708
+ api: "anthropic-messages",
11709
+ provider: "vercel-ai-gateway",
11710
+ baseUrl: "https://ai-gateway.vercel.sh",
11711
+ reasoning: false,
11712
+ input: ["text", "image"],
11713
+ cost: {
11714
+ input: 0.14,
11715
+ output: 0.39999999999999997,
11716
+ cacheRead: 0,
11717
+ cacheWrite: 0,
11718
+ },
11719
+ contextWindow: 262144,
11720
+ maxTokens: 131072,
11721
+ },
11603
11722
  "inception/mercury-2": {
11604
11723
  id: "inception/mercury-2",
11605
11724
  name: "Mercury 2",
@@ -11668,23 +11787,6 @@ export const MODELS = {
11668
11787
  contextWindow: 128000,
11669
11788
  maxTokens: 8192,
11670
11789
  },
11671
- "meituan/longcat-flash-thinking": {
11672
- id: "meituan/longcat-flash-thinking",
11673
- name: "LongCat Flash Thinking",
11674
- api: "anthropic-messages",
11675
- provider: "vercel-ai-gateway",
11676
- baseUrl: "https://ai-gateway.vercel.sh",
11677
- reasoning: true,
11678
- input: ["text"],
11679
- cost: {
11680
- input: 0.15,
11681
- output: 1.5,
11682
- cacheRead: 0,
11683
- cacheWrite: 0,
11684
- },
11685
- contextWindow: 128000,
11686
- maxTokens: 8192,
11687
- },
11688
11790
  "meta/llama-3.1-70b": {
11689
11791
  id: "meta/llama-3.1-70b",
11690
11792
  name: "Llama 3.1 70B Instruct",
@@ -11779,12 +11881,12 @@ export const MODELS = {
11779
11881
  reasoning: false,
11780
11882
  input: ["text", "image"],
11781
11883
  cost: {
11782
- input: 0.35,
11783
- output: 1.15,
11884
+ input: 0.24,
11885
+ output: 0.9700000000000001,
11784
11886
  cacheRead: 0,
11785
11887
  cacheWrite: 0,
11786
11888
  },
11787
- contextWindow: 524288,
11889
+ contextWindow: 128000,
11788
11890
  maxTokens: 8192,
11789
11891
  },
11790
11892
  "meta/llama-4-scout": {
@@ -12095,20 +12197,20 @@ export const MODELS = {
12095
12197
  },
12096
12198
  "moonshotai/kimi-k2": {
12097
12199
  id: "moonshotai/kimi-k2",
12098
- name: "Kimi K2",
12200
+ name: "Kimi K2 Instruct",
12099
12201
  api: "anthropic-messages",
12100
12202
  provider: "vercel-ai-gateway",
12101
12203
  baseUrl: "https://ai-gateway.vercel.sh",
12102
12204
  reasoning: false,
12103
12205
  input: ["text"],
12104
12206
  cost: {
12105
- input: 0.6,
12106
- output: 2.5,
12107
- cacheRead: 0.15,
12207
+ input: 0.5700000000000001,
12208
+ output: 2.3,
12209
+ cacheRead: 0,
12108
12210
  cacheWrite: 0,
12109
12211
  },
12110
12212
  contextWindow: 131072,
12111
- maxTokens: 16384,
12213
+ maxTokens: 131072,
12112
12214
  },
12113
12215
  "moonshotai/kimi-k2-0905": {
12114
12216
  id: "moonshotai/kimi-k2-0905",
@@ -12121,11 +12223,11 @@ export const MODELS = {
12121
12223
  cost: {
12122
12224
  input: 0.6,
12123
12225
  output: 2.5,
12124
- cacheRead: 0.15,
12226
+ cacheRead: 0.3,
12125
12227
  cacheWrite: 0,
12126
12228
  },
12127
12229
  contextWindow: 256000,
12128
- maxTokens: 16384,
12230
+ maxTokens: 128000,
12129
12231
  },
12130
12232
  "moonshotai/kimi-k2-thinking": {
12131
12233
  id: "moonshotai/kimi-k2-thinking",
@@ -12714,12 +12816,12 @@ export const MODELS = {
12714
12816
  reasoning: true,
12715
12817
  input: ["text"],
12716
12818
  cost: {
12717
- input: 0.07,
12718
- output: 0.3,
12819
+ input: 0.049999999999999996,
12820
+ output: 0.19999999999999998,
12719
12821
  cacheRead: 0,
12720
12822
  cacheWrite: 0,
12721
12823
  },
12722
- contextWindow: 128000,
12824
+ contextWindow: 131072,
12723
12825
  maxTokens: 8192,
12724
12826
  },
12725
12827
  "openai/gpt-oss-safeguard-20b": {
@@ -13402,6 +13504,23 @@ export const MODELS = {
13402
13504
  contextWindow: 202800,
13403
13505
  maxTokens: 131100,
13404
13506
  },
13507
+ "zai/glm-5v-turbo": {
13508
+ id: "zai/glm-5v-turbo",
13509
+ name: "GLM 5V Turbo",
13510
+ api: "anthropic-messages",
13511
+ provider: "vercel-ai-gateway",
13512
+ baseUrl: "https://ai-gateway.vercel.sh",
13513
+ reasoning: true,
13514
+ input: ["text", "image"],
13515
+ cost: {
13516
+ input: 1.2,
13517
+ output: 4,
13518
+ cacheRead: 0.24,
13519
+ cacheWrite: 0,
13520
+ },
13521
+ contextWindow: 200000,
13522
+ maxTokens: 128000,
13523
+ },
13405
13524
  },
13406
13525
  "xai": {
13407
13526
  "grok-2": {
@@ -13824,9 +13943,9 @@ export const MODELS = {
13824
13943
  reasoning: true,
13825
13944
  input: ["text"],
13826
13945
  cost: {
13827
- input: 0.6,
13828
- output: 2.2,
13829
- cacheRead: 0.11,
13946
+ input: 0,
13947
+ output: 0,
13948
+ cacheRead: 0,
13830
13949
  cacheWrite: 0,
13831
13950
  },
13832
13951
  contextWindow: 131072,
@@ -13842,9 +13961,9 @@ export const MODELS = {
13842
13961
  reasoning: true,
13843
13962
  input: ["text"],
13844
13963
  cost: {
13845
- input: 0.2,
13846
- output: 1.1,
13847
- cacheRead: 0.03,
13964
+ input: 0,
13965
+ output: 0,
13966
+ cacheRead: 0,
13848
13967
  cacheWrite: 0,
13849
13968
  },
13850
13969
  contextWindow: 131072,
@@ -13878,8 +13997,8 @@ export const MODELS = {
13878
13997
  reasoning: true,
13879
13998
  input: ["text", "image"],
13880
13999
  cost: {
13881
- input: 0.6,
13882
- output: 1.8,
14000
+ input: 0,
14001
+ output: 0,
13883
14002
  cacheRead: 0,
13884
14003
  cacheWrite: 0,
13885
14004
  },
@@ -13892,13 +14011,13 @@ export const MODELS = {
13892
14011
  api: "openai-completions",
13893
14012
  provider: "zai",
13894
14013
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13895
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14014
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13896
14015
  reasoning: true,
13897
14016
  input: ["text"],
13898
14017
  cost: {
13899
- input: 0.6,
13900
- output: 2.2,
13901
- cacheRead: 0.11,
14018
+ input: 0,
14019
+ output: 0,
14020
+ cacheRead: 0,
13902
14021
  cacheWrite: 0,
13903
14022
  },
13904
14023
  contextWindow: 204800,
@@ -13910,12 +14029,12 @@ export const MODELS = {
13910
14029
  api: "openai-completions",
13911
14030
  provider: "zai",
13912
14031
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13913
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14032
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13914
14033
  reasoning: true,
13915
14034
  input: ["text", "image"],
13916
14035
  cost: {
13917
- input: 0.3,
13918
- output: 0.9,
14036
+ input: 0,
14037
+ output: 0,
13919
14038
  cacheRead: 0,
13920
14039
  cacheWrite: 0,
13921
14040
  },
@@ -13928,13 +14047,13 @@ export const MODELS = {
13928
14047
  api: "openai-completions",
13929
14048
  provider: "zai",
13930
14049
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13931
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14050
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13932
14051
  reasoning: true,
13933
14052
  input: ["text"],
13934
14053
  cost: {
13935
- input: 0.6,
13936
- output: 2.2,
13937
- cacheRead: 0.11,
14054
+ input: 0,
14055
+ output: 0,
14056
+ cacheRead: 0,
13938
14057
  cacheWrite: 0,
13939
14058
  },
13940
14059
  contextWindow: 204800,
@@ -13946,7 +14065,7 @@ export const MODELS = {
13946
14065
  api: "openai-completions",
13947
14066
  provider: "zai",
13948
14067
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13949
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14068
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13950
14069
  reasoning: true,
13951
14070
  input: ["text"],
13952
14071
  cost: {
@@ -13964,7 +14083,7 @@ export const MODELS = {
13964
14083
  api: "openai-completions",
13965
14084
  provider: "zai",
13966
14085
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13967
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14086
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13968
14087
  reasoning: true,
13969
14088
  input: ["text"],
13970
14089
  cost: {
@@ -13982,13 +14101,13 @@ export const MODELS = {
13982
14101
  api: "openai-completions",
13983
14102
  provider: "zai",
13984
14103
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13985
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14104
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13986
14105
  reasoning: true,
13987
14106
  input: ["text"],
13988
14107
  cost: {
13989
- input: 1,
13990
- output: 3.2,
13991
- cacheRead: 0.2,
14108
+ input: 0,
14109
+ output: 0,
14110
+ cacheRead: 0,
13992
14111
  cacheWrite: 0,
13993
14112
  },
13994
14113
  contextWindow: 204800,
@@ -14000,13 +14119,49 @@ export const MODELS = {
14000
14119
  api: "openai-completions",
14001
14120
  provider: "zai",
14002
14121
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
14003
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14122
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14004
14123
  reasoning: true,
14005
14124
  input: ["text"],
14006
14125
  cost: {
14007
- input: 1.2,
14008
- output: 4,
14009
- cacheRead: 0.24,
14126
+ input: 0,
14127
+ output: 0,
14128
+ cacheRead: 0,
14129
+ cacheWrite: 0,
14130
+ },
14131
+ contextWindow: 200000,
14132
+ maxTokens: 131072,
14133
+ },
14134
+ "glm-5.1": {
14135
+ id: "glm-5.1",
14136
+ name: "GLM-5.1",
14137
+ api: "openai-completions",
14138
+ provider: "zai",
14139
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
14140
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14141
+ reasoning: true,
14142
+ input: ["text"],
14143
+ cost: {
14144
+ input: 0,
14145
+ output: 0,
14146
+ cacheRead: 0,
14147
+ cacheWrite: 0,
14148
+ },
14149
+ contextWindow: 200000,
14150
+ maxTokens: 131072,
14151
+ },
14152
+ "glm-5v-turbo": {
14153
+ id: "glm-5v-turbo",
14154
+ name: "glm-5v-turbo",
14155
+ api: "openai-completions",
14156
+ provider: "zai",
14157
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
14158
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14159
+ reasoning: true,
14160
+ input: ["text", "image"],
14161
+ cost: {
14162
+ input: 0,
14163
+ output: 0,
14164
+ cacheRead: 0,
14010
14165
  cacheWrite: 0,
14011
14166
  },
14012
14167
  contextWindow: 200000,