omk-ai 0.95.0 → 0.95.2

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.
Files changed (46) hide show
  1. package/README.md +31 -2
  2. package/dist/index.d.ts +1 -0
  3. package/dist/index.d.ts.map +1 -1
  4. package/dist/index.js +1 -0
  5. package/dist/index.js.map +1 -1
  6. package/dist/models.generated.d.ts +236 -681
  7. package/dist/models.generated.d.ts.map +1 -1
  8. package/dist/models.generated.js +81 -622
  9. package/dist/models.generated.js.map +1 -1
  10. package/dist/providers/anthropic.d.ts.map +1 -1
  11. package/dist/providers/anthropic.js +45 -20
  12. package/dist/providers/anthropic.js.map +1 -1
  13. package/dist/providers/azure-openai-responses.d.ts.map +1 -1
  14. package/dist/providers/azure-openai-responses.js +2 -2
  15. package/dist/providers/azure-openai-responses.js.map +1 -1
  16. package/dist/providers/openai-codex-moa-stream-limits.d.ts +0 -1
  17. package/dist/providers/openai-codex-moa-stream-limits.d.ts.map +1 -1
  18. package/dist/providers/openai-codex-moa-stream-limits.js +1 -17
  19. package/dist/providers/openai-codex-moa-stream-limits.js.map +1 -1
  20. package/dist/providers/openai-codex-moa.d.ts.map +1 -1
  21. package/dist/providers/openai-codex-moa.js +2 -11
  22. package/dist/providers/openai-codex-moa.js.map +1 -1
  23. package/dist/providers/openai-codex-responses.d.ts.map +1 -1
  24. package/dist/providers/openai-codex-responses.js +2 -2
  25. package/dist/providers/openai-codex-responses.js.map +1 -1
  26. package/dist/providers/openai-completions.d.ts.map +1 -1
  27. package/dist/providers/openai-completions.js +23 -10
  28. package/dist/providers/openai-completions.js.map +1 -1
  29. package/dist/providers/openai-prompt-cache.d.ts +6 -0
  30. package/dist/providers/openai-prompt-cache.d.ts.map +1 -1
  31. package/dist/providers/openai-prompt-cache.js +13 -0
  32. package/dist/providers/openai-prompt-cache.js.map +1 -1
  33. package/dist/providers/openai-responses.d.ts.map +1 -1
  34. package/dist/providers/openai-responses.js +4 -2
  35. package/dist/providers/openai-responses.js.map +1 -1
  36. package/dist/providers/prompt-cache.d.ts +10 -0
  37. package/dist/providers/prompt-cache.d.ts.map +1 -0
  38. package/dist/providers/prompt-cache.js +40 -0
  39. package/dist/providers/prompt-cache.js.map +1 -0
  40. package/dist/providers/tool-schema.d.ts.map +1 -1
  41. package/dist/providers/tool-schema.js +10 -4
  42. package/dist/providers/tool-schema.js.map +1 -1
  43. package/dist/types.d.ts +7 -0
  44. package/dist/types.d.ts.map +1 -1
  45. package/dist/types.js.map +1 -1
  46. package/package.json +5 -5
@@ -3832,6 +3832,7 @@ export const MODELS = {
3832
3832
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3833
3833
  compat: { "sendSessionAffinityHeaders": true },
3834
3834
  reasoning: true,
3835
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
3835
3836
  input: ["text"],
3836
3837
  cost: {
3837
3838
  input: 1.4,
@@ -4069,6 +4070,7 @@ export const MODELS = {
4069
4070
  baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
4070
4071
  compat: { "sendSessionAffinityHeaders": true },
4071
4072
  reasoning: true,
4073
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
4072
4074
  input: ["text"],
4073
4075
  cost: {
4074
4076
  input: 1.4,
@@ -4165,6 +4167,7 @@ export const MODELS = {
4165
4167
  baseUrl: "https://api.fireworks.ai/inference",
4166
4168
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4167
4169
  reasoning: true,
4170
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
4168
4171
  input: ["text"],
4169
4172
  cost: {
4170
4173
  input: 1.4,
@@ -4328,6 +4331,7 @@ export const MODELS = {
4328
4331
  baseUrl: "https://api.fireworks.ai/inference",
4329
4332
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4330
4333
  reasoning: true,
4334
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
4331
4335
  input: ["text"],
4332
4336
  cost: {
4333
4337
  input: 2.1,
@@ -6694,6 +6698,7 @@ export const MODELS = {
6694
6698
  baseUrl: "https://router.huggingface.co/v1",
6695
6699
  compat: { "supportsDeveloperRole": false },
6696
6700
  reasoning: true,
6701
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
6697
6702
  input: ["text"],
6698
6703
  cost: {
6699
6704
  input: 1.4,
@@ -8326,6 +8331,7 @@ export const MODELS = {
8326
8331
  headers: { "NVCF-POLL-SECONDS": "3600" },
8327
8332
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
8328
8333
  reasoning: true,
8334
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
8329
8335
  input: ["text"],
8330
8336
  cost: {
8331
8337
  input: 0,
@@ -9625,6 +9631,7 @@ export const MODELS = {
9625
9631
  provider: "opencode",
9626
9632
  baseUrl: "https://opencode.ai/zen/v1",
9627
9633
  reasoning: true,
9634
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
9628
9635
  input: ["text"],
9629
9636
  cost: {
9630
9637
  input: 1.4,
@@ -10318,6 +10325,7 @@ export const MODELS = {
10318
10325
  provider: "opencode-go",
10319
10326
  baseUrl: "https://opencode.ai/zen/go/v1",
10320
10327
  reasoning: true,
10328
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
10321
10329
  input: ["text"],
10322
10330
  cost: {
10323
10331
  input: 1.4,
@@ -10745,23 +10753,6 @@ export const MODELS = {
10745
10753
  contextWindow: 1000000,
10746
10754
  maxTokens: 128000,
10747
10755
  },
10748
- "anthropic/claude-fable-5:batch": {
10749
- id: "anthropic/claude-fable-5:batch",
10750
- name: "Anthropic: Claude Fable 5 (batch)",
10751
- api: "openai-completions",
10752
- provider: "openrouter",
10753
- baseUrl: "https://openrouter.ai/api/v1",
10754
- reasoning: true,
10755
- input: ["text", "image"],
10756
- cost: {
10757
- input: 5,
10758
- output: 25,
10759
- cacheRead: 0.5,
10760
- cacheWrite: 6.25,
10761
- },
10762
- contextWindow: 1000000,
10763
- maxTokens: 128000,
10764
- },
10765
10756
  "anthropic/claude-haiku-4.5": {
10766
10757
  id: "anthropic/claude-haiku-4.5",
10767
10758
  name: "Anthropic: Claude Haiku 4.5",
@@ -10779,23 +10770,6 @@ export const MODELS = {
10779
10770
  contextWindow: 200000,
10780
10771
  maxTokens: 64000,
10781
10772
  },
10782
- "anthropic/claude-haiku-4.5:batch": {
10783
- id: "anthropic/claude-haiku-4.5:batch",
10784
- name: "Anthropic: Claude Haiku 4.5 (batch)",
10785
- api: "openai-completions",
10786
- provider: "openrouter",
10787
- baseUrl: "https://openrouter.ai/api/v1",
10788
- reasoning: true,
10789
- input: ["text", "image"],
10790
- cost: {
10791
- input: 0.5,
10792
- output: 2.5,
10793
- cacheRead: 0.049999999999999996,
10794
- cacheWrite: 0.625,
10795
- },
10796
- contextWindow: 200000,
10797
- maxTokens: 64000,
10798
- },
10799
10773
  "anthropic/claude-opus-4": {
10800
10774
  id: "anthropic/claude-opus-4",
10801
10775
  name: "Anthropic: Claude Opus 4",
@@ -10830,23 +10804,6 @@ export const MODELS = {
10830
10804
  contextWindow: 200000,
10831
10805
  maxTokens: 32000,
10832
10806
  },
10833
- "anthropic/claude-opus-4.1:batch": {
10834
- id: "anthropic/claude-opus-4.1:batch",
10835
- name: "Anthropic: Claude Opus 4.1 (batch)",
10836
- api: "openai-completions",
10837
- provider: "openrouter",
10838
- baseUrl: "https://openrouter.ai/api/v1",
10839
- reasoning: true,
10840
- input: ["text", "image"],
10841
- cost: {
10842
- input: 7.5,
10843
- output: 37.5,
10844
- cacheRead: 0.75,
10845
- cacheWrite: 9.375,
10846
- },
10847
- contextWindow: 200000,
10848
- maxTokens: 32000,
10849
- },
10850
10807
  "anthropic/claude-opus-4.5": {
10851
10808
  id: "anthropic/claude-opus-4.5",
10852
10809
  name: "Anthropic: Claude Opus 4.5",
@@ -10864,23 +10821,6 @@ export const MODELS = {
10864
10821
  contextWindow: 200000,
10865
10822
  maxTokens: 64000,
10866
10823
  },
10867
- "anthropic/claude-opus-4.5:batch": {
10868
- id: "anthropic/claude-opus-4.5:batch",
10869
- name: "Anthropic: Claude Opus 4.5 (batch)",
10870
- api: "openai-completions",
10871
- provider: "openrouter",
10872
- baseUrl: "https://openrouter.ai/api/v1",
10873
- reasoning: true,
10874
- input: ["text", "image"],
10875
- cost: {
10876
- input: 2.5,
10877
- output: 12.5,
10878
- cacheRead: 0.25,
10879
- cacheWrite: 3.125,
10880
- },
10881
- contextWindow: 200000,
10882
- maxTokens: 64000,
10883
- },
10884
10824
  "anthropic/claude-opus-4.6": {
10885
10825
  id: "anthropic/claude-opus-4.6",
10886
10826
  name: "Anthropic: Claude Opus 4.6",
@@ -10899,24 +10839,6 @@ export const MODELS = {
10899
10839
  contextWindow: 1000000,
10900
10840
  maxTokens: 128000,
10901
10841
  },
10902
- "anthropic/claude-opus-4.6:batch": {
10903
- id: "anthropic/claude-opus-4.6:batch",
10904
- name: "Anthropic: Claude Opus 4.6 (batch)",
10905
- api: "openai-completions",
10906
- provider: "openrouter",
10907
- baseUrl: "https://openrouter.ai/api/v1",
10908
- reasoning: true,
10909
- thinkingLevelMap: { "xhigh": "max" },
10910
- input: ["text", "image"],
10911
- cost: {
10912
- input: 2.5,
10913
- output: 12.5,
10914
- cacheRead: 0.25,
10915
- cacheWrite: 3.125,
10916
- },
10917
- contextWindow: 1000000,
10918
- maxTokens: 128000,
10919
- },
10920
10842
  "anthropic/claude-opus-4.7": {
10921
10843
  id: "anthropic/claude-opus-4.7",
10922
10844
  name: "Anthropic: Claude Opus 4.7",
@@ -10953,24 +10875,6 @@ export const MODELS = {
10953
10875
  contextWindow: 1000000,
10954
10876
  maxTokens: 128000,
10955
10877
  },
10956
- "anthropic/claude-opus-4.7:batch": {
10957
- id: "anthropic/claude-opus-4.7:batch",
10958
- name: "Anthropic: Claude Opus 4.7 (batch)",
10959
- api: "openai-completions",
10960
- provider: "openrouter",
10961
- baseUrl: "https://openrouter.ai/api/v1",
10962
- reasoning: true,
10963
- thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
10964
- input: ["text", "image"],
10965
- cost: {
10966
- input: 2.5,
10967
- output: 12.5,
10968
- cacheRead: 0.25,
10969
- cacheWrite: 3.125,
10970
- },
10971
- contextWindow: 1000000,
10972
- maxTokens: 128000,
10973
- },
10974
10878
  "anthropic/claude-opus-4.8": {
10975
10879
  id: "anthropic/claude-opus-4.8",
10976
10880
  name: "Anthropic: Claude Opus 4.8",
@@ -11007,24 +10911,6 @@ export const MODELS = {
11007
10911
  contextWindow: 1000000,
11008
10912
  maxTokens: 128000,
11009
10913
  },
11010
- "anthropic/claude-opus-4.8:batch": {
11011
- id: "anthropic/claude-opus-4.8:batch",
11012
- name: "Anthropic: Claude Opus 4.8 (batch)",
11013
- api: "openai-completions",
11014
- provider: "openrouter",
11015
- baseUrl: "https://openrouter.ai/api/v1",
11016
- reasoning: true,
11017
- thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
11018
- input: ["text", "image"],
11019
- cost: {
11020
- input: 2.5,
11021
- output: 12.5,
11022
- cacheRead: 0.25,
11023
- cacheWrite: 3.125,
11024
- },
11025
- contextWindow: 1000000,
11026
- maxTokens: 128000,
11027
- },
11028
10914
  "anthropic/claude-opus-5": {
11029
10915
  id: "anthropic/claude-opus-5",
11030
10916
  name: "Claude Opus 5",
@@ -11093,23 +10979,6 @@ export const MODELS = {
11093
10979
  contextWindow: 1000000,
11094
10980
  maxTokens: 64000,
11095
10981
  },
11096
- "anthropic/claude-sonnet-4.5:batch": {
11097
- id: "anthropic/claude-sonnet-4.5:batch",
11098
- name: "Anthropic: Claude Sonnet 4.5 (batch)",
11099
- api: "openai-completions",
11100
- provider: "openrouter",
11101
- baseUrl: "https://openrouter.ai/api/v1",
11102
- reasoning: true,
11103
- input: ["text", "image"],
11104
- cost: {
11105
- input: 1.5,
11106
- output: 7.5,
11107
- cacheRead: 0.15,
11108
- cacheWrite: 1.875,
11109
- },
11110
- contextWindow: 1000000,
11111
- maxTokens: 64000,
11112
- },
11113
10982
  "anthropic/claude-sonnet-4.6": {
11114
10983
  id: "anthropic/claude-sonnet-4.6",
11115
10984
  name: "Anthropic: Claude Sonnet 4.6",
@@ -11145,23 +11014,6 @@ export const MODELS = {
11145
11014
  contextWindow: 1000000,
11146
11015
  maxTokens: 128000,
11147
11016
  },
11148
- "anthropic/claude-sonnet-5:batch": {
11149
- id: "anthropic/claude-sonnet-5:batch",
11150
- name: "Anthropic: Claude Sonnet 5 (batch)",
11151
- api: "openai-completions",
11152
- provider: "openrouter",
11153
- baseUrl: "https://openrouter.ai/api/v1",
11154
- reasoning: true,
11155
- input: ["text", "image"],
11156
- cost: {
11157
- input: 1,
11158
- output: 5,
11159
- cacheRead: 0.09999999999999999,
11160
- cacheWrite: 1.25,
11161
- },
11162
- contextWindow: 1000000,
11163
- maxTokens: 128000,
11164
- },
11165
11017
  "arcee-ai/trinity-large-thinking": {
11166
11018
  id: "arcee-ai/trinity-large-thinking",
11167
11019
  name: "Arcee AI: Trinity Large Thinking",
@@ -11559,40 +11411,6 @@ export const MODELS = {
11559
11411
  contextWindow: 1048576,
11560
11412
  maxTokens: 65535,
11561
11413
  },
11562
- "google/gemini-2.5-flash-lite:batch": {
11563
- id: "google/gemini-2.5-flash-lite:batch",
11564
- name: "Google: Gemini 2.5 Flash Lite (batch)",
11565
- api: "openai-completions",
11566
- provider: "openrouter",
11567
- baseUrl: "https://openrouter.ai/api/v1",
11568
- reasoning: true,
11569
- input: ["text", "image"],
11570
- cost: {
11571
- input: 0.049999999999999996,
11572
- output: 0.19999999999999998,
11573
- cacheRead: 0.01,
11574
- cacheWrite: 0,
11575
- },
11576
- contextWindow: 1048576,
11577
- maxTokens: 65535,
11578
- },
11579
- "google/gemini-2.5-flash:batch": {
11580
- id: "google/gemini-2.5-flash:batch",
11581
- name: "Google: Gemini 2.5 Flash (batch)",
11582
- api: "openai-completions",
11583
- provider: "openrouter",
11584
- baseUrl: "https://openrouter.ai/api/v1",
11585
- reasoning: true,
11586
- input: ["text", "image"],
11587
- cost: {
11588
- input: 0.15,
11589
- output: 1.25,
11590
- cacheRead: 0.03,
11591
- cacheWrite: 0,
11592
- },
11593
- contextWindow: 1048576,
11594
- maxTokens: 65535,
11595
- },
11596
11414
  "google/gemini-2.5-pro": {
11597
11415
  id: "google/gemini-2.5-pro",
11598
11416
  name: "Google: Gemini 2.5 Pro",
@@ -11644,23 +11462,6 @@ export const MODELS = {
11644
11462
  contextWindow: 1048576,
11645
11463
  maxTokens: 65535,
11646
11464
  },
11647
- "google/gemini-2.5-pro:batch": {
11648
- id: "google/gemini-2.5-pro:batch",
11649
- name: "Google: Gemini 2.5 Pro (batch)",
11650
- api: "openai-completions",
11651
- provider: "openrouter",
11652
- baseUrl: "https://openrouter.ai/api/v1",
11653
- reasoning: true,
11654
- input: ["text", "image"],
11655
- cost: {
11656
- input: 0.625,
11657
- output: 5,
11658
- cacheRead: 0.125,
11659
- cacheWrite: 0,
11660
- },
11661
- contextWindow: 1048576,
11662
- maxTokens: 65536,
11663
- },
11664
11465
  "google/gemini-3-flash-preview": {
11665
11466
  id: "google/gemini-3-flash-preview",
11666
11467
  name: "Google: Gemini 3 Flash Preview",
@@ -11678,23 +11479,6 @@ export const MODELS = {
11678
11479
  contextWindow: 1048576,
11679
11480
  maxTokens: 65535,
11680
11481
  },
11681
- "google/gemini-3-flash-preview:batch": {
11682
- id: "google/gemini-3-flash-preview:batch",
11683
- name: "Google: Gemini 3 Flash Preview (batch)",
11684
- api: "openai-completions",
11685
- provider: "openrouter",
11686
- baseUrl: "https://openrouter.ai/api/v1",
11687
- reasoning: true,
11688
- input: ["text", "image"],
11689
- cost: {
11690
- input: 0.25,
11691
- output: 1.5,
11692
- cacheRead: 0,
11693
- cacheWrite: 0,
11694
- },
11695
- contextWindow: 1048576,
11696
- maxTokens: 65535,
11697
- },
11698
11482
  "google/gemini-3-pro-image": {
11699
11483
  id: "google/gemini-3-pro-image",
11700
11484
  name: "Google: Nano Banana Pro (Gemini 3 Pro Image)",
@@ -11746,23 +11530,6 @@ export const MODELS = {
11746
11530
  contextWindow: 1048576,
11747
11531
  maxTokens: 65536,
11748
11532
  },
11749
- "google/gemini-3.1-flash-lite:batch": {
11750
- id: "google/gemini-3.1-flash-lite:batch",
11751
- name: "Google: Gemini 3.1 Flash Lite (batch)",
11752
- api: "openai-completions",
11753
- provider: "openrouter",
11754
- baseUrl: "https://openrouter.ai/api/v1",
11755
- reasoning: true,
11756
- input: ["text", "image"],
11757
- cost: {
11758
- input: 0.125,
11759
- output: 0.75,
11760
- cacheRead: 0.012499999999999999,
11761
- cacheWrite: 0,
11762
- },
11763
- contextWindow: 1048576,
11764
- maxTokens: 65536,
11765
- },
11766
11533
  "google/gemini-3.1-pro-preview": {
11767
11534
  id: "google/gemini-3.1-pro-preview",
11768
11535
  name: "Google: Gemini 3.1 Pro Preview",
@@ -11797,23 +11564,6 @@ export const MODELS = {
11797
11564
  contextWindow: 1048576,
11798
11565
  maxTokens: 65536,
11799
11566
  },
11800
- "google/gemini-3.1-pro-preview:batch": {
11801
- id: "google/gemini-3.1-pro-preview:batch",
11802
- name: "Google: Gemini 3.1 Pro Preview (batch)",
11803
- api: "openai-completions",
11804
- provider: "openrouter",
11805
- baseUrl: "https://openrouter.ai/api/v1",
11806
- reasoning: true,
11807
- input: ["text", "image"],
11808
- cost: {
11809
- input: 1,
11810
- output: 6,
11811
- cacheRead: 0,
11812
- cacheWrite: 0,
11813
- },
11814
- contextWindow: 1048576,
11815
- maxTokens: 65536,
11816
- },
11817
11567
  "google/gemini-3.5-flash": {
11818
11568
  id: "google/gemini-3.5-flash",
11819
11569
  name: "Google: Gemini 3.5 Flash",
@@ -11848,40 +11598,6 @@ export const MODELS = {
11848
11598
  contextWindow: 1048576,
11849
11599
  maxTokens: 65536,
11850
11600
  },
11851
- "google/gemini-3.5-flash-lite:batch": {
11852
- id: "google/gemini-3.5-flash-lite:batch",
11853
- name: "Google: Gemini 3.5 Flash Lite (batch)",
11854
- api: "openai-completions",
11855
- provider: "openrouter",
11856
- baseUrl: "https://openrouter.ai/api/v1",
11857
- reasoning: true,
11858
- input: ["text", "image"],
11859
- cost: {
11860
- input: 0.15,
11861
- output: 1.25,
11862
- cacheRead: 0.015,
11863
- cacheWrite: 0,
11864
- },
11865
- contextWindow: 1048576,
11866
- maxTokens: 65536,
11867
- },
11868
- "google/gemini-3.5-flash:batch": {
11869
- id: "google/gemini-3.5-flash:batch",
11870
- name: "Google: Gemini 3.5 Flash (batch)",
11871
- api: "openai-completions",
11872
- provider: "openrouter",
11873
- baseUrl: "https://openrouter.ai/api/v1",
11874
- reasoning: true,
11875
- input: ["text", "image"],
11876
- cost: {
11877
- input: 0.75,
11878
- output: 4.5,
11879
- cacheRead: 0.075,
11880
- cacheWrite: 0,
11881
- },
11882
- contextWindow: 1048576,
11883
- maxTokens: 65536,
11884
- },
11885
11601
  "google/gemini-3.6-flash": {
11886
11602
  id: "google/gemini-3.6-flash",
11887
11603
  name: "Google: Gemini 3.6 Flash",
@@ -11899,23 +11615,6 @@ export const MODELS = {
11899
11615
  contextWindow: 1048576,
11900
11616
  maxTokens: 65536,
11901
11617
  },
11902
- "google/gemini-3.6-flash:batch": {
11903
- id: "google/gemini-3.6-flash:batch",
11904
- name: "Google: Gemini 3.6 Flash (batch)",
11905
- api: "openai-completions",
11906
- provider: "openrouter",
11907
- baseUrl: "https://openrouter.ai/api/v1",
11908
- reasoning: true,
11909
- input: ["text", "image"],
11910
- cost: {
11911
- input: 0.75,
11912
- output: 3.75,
11913
- cacheRead: 0.075,
11914
- cacheWrite: 0.08333333333333334,
11915
- },
11916
- contextWindow: 1048576,
11917
- maxTokens: 65536,
11918
- },
11919
11618
  "google/gemma-3-12b-it": {
11920
11619
  id: "google/gemma-3-12b-it",
11921
11620
  name: "Google: Gemma 3 12B",
@@ -12393,23 +12092,6 @@ export const MODELS = {
12393
12092
  contextWindow: 1048576,
12394
12093
  maxTokens: 512000,
12395
12094
  },
12396
- "minimax/minimax-m3:batch": {
12397
- id: "minimax/minimax-m3:batch",
12398
- name: "MiniMax: MiniMax M3 (batch)",
12399
- api: "openai-completions",
12400
- provider: "openrouter",
12401
- baseUrl: "https://openrouter.ai/api/v1",
12402
- reasoning: true,
12403
- input: ["text", "image"],
12404
- cost: {
12405
- input: 0.15,
12406
- output: 0.6,
12407
- cacheRead: 0.03,
12408
- cacheWrite: 0,
12409
- },
12410
- contextWindow: 524288,
12411
- maxTokens: 4096,
12412
- },
12413
12095
  "mistralai/codestral-2508": {
12414
12096
  id: "mistralai/codestral-2508",
12415
12097
  name: "Mistral: Codestral 2508",
@@ -12427,23 +12109,6 @@ export const MODELS = {
12427
12109
  contextWindow: 256000,
12428
12110
  maxTokens: 4096,
12429
12111
  },
12430
- "mistralai/devstral-2512": {
12431
- id: "mistralai/devstral-2512",
12432
- name: "Mistral: Devstral 2 2512",
12433
- api: "openai-completions",
12434
- provider: "openrouter",
12435
- baseUrl: "https://openrouter.ai/api/v1",
12436
- reasoning: false,
12437
- input: ["text"],
12438
- cost: {
12439
- input: 0.39999999999999997,
12440
- output: 2,
12441
- cacheRead: 0.04,
12442
- cacheWrite: 0,
12443
- },
12444
- contextWindow: 262144,
12445
- maxTokens: 4096,
12446
- },
12447
12112
  "mistralai/ministral-14b-2512": {
12448
12113
  id: "mistralai/ministral-14b-2512",
12449
12114
  name: "Mistral: Ministral 3 14B 2512",
@@ -12657,13 +12322,13 @@ export const MODELS = {
12657
12322
  reasoning: false,
12658
12323
  input: ["text", "image"],
12659
12324
  cost: {
12660
- input: 0.09999999999999999,
12661
- output: 0.3,
12662
- cacheRead: 0.01,
12325
+ input: 0.075,
12326
+ output: 0.19999999999999998,
12327
+ cacheRead: 0,
12663
12328
  cacheWrite: 0,
12664
12329
  },
12665
12330
  contextWindow: 256000,
12666
- maxTokens: 4096,
12331
+ maxTokens: 16384,
12667
12332
  },
12668
12333
  "mistralai/mixtral-8x22b-instruct": {
12669
12334
  id: "mistralai/mixtral-8x22b-instruct",
@@ -12777,9 +12442,9 @@ export const MODELS = {
12777
12442
  reasoning: true,
12778
12443
  input: ["text", "image"],
12779
12444
  cost: {
12780
- input: 0.95,
12781
- output: 4,
12782
- cacheRead: 0.16,
12445
+ input: 0.6,
12446
+ output: 3.41,
12447
+ cacheRead: 0.19999999999999998,
12783
12448
  cacheWrite: 0,
12784
12449
  },
12785
12450
  contextWindow: 262144,
@@ -13295,23 +12960,6 @@ export const MODELS = {
13295
12960
  contextWindow: 400000,
13296
12961
  maxTokens: 128000,
13297
12962
  },
13298
- "openai/gpt-5-mini:batch": {
13299
- id: "openai/gpt-5-mini:batch",
13300
- name: "OpenAI: GPT-5 Mini (batch)",
13301
- api: "openai-completions",
13302
- provider: "openrouter",
13303
- baseUrl: "https://openrouter.ai/api/v1",
13304
- reasoning: true,
13305
- input: ["text", "image"],
13306
- cost: {
13307
- input: 0.125,
13308
- output: 1,
13309
- cacheRead: 0.012499999999999999,
13310
- cacheWrite: 0,
13311
- },
13312
- contextWindow: 400000,
13313
- maxTokens: 128000,
13314
- },
13315
12963
  "openai/gpt-5-nano": {
13316
12964
  id: "openai/gpt-5-nano",
13317
12965
  name: "OpenAI: GPT-5 Nano",
@@ -13329,23 +12977,6 @@ export const MODELS = {
13329
12977
  contextWindow: 400000,
13330
12978
  maxTokens: 128000,
13331
12979
  },
13332
- "openai/gpt-5-nano:batch": {
13333
- id: "openai/gpt-5-nano:batch",
13334
- name: "OpenAI: GPT-5 Nano (batch)",
13335
- api: "openai-completions",
13336
- provider: "openrouter",
13337
- baseUrl: "https://openrouter.ai/api/v1",
13338
- reasoning: true,
13339
- input: ["text", "image"],
13340
- cost: {
13341
- input: 0.024999999999999998,
13342
- output: 0.19999999999999998,
13343
- cacheRead: 0.0025,
13344
- cacheWrite: 0,
13345
- },
13346
- contextWindow: 400000,
13347
- maxTokens: 128000,
13348
- },
13349
12980
  "openai/gpt-5-pro": {
13350
12981
  id: "openai/gpt-5-pro",
13351
12982
  name: "OpenAI: GPT-5 Pro",
@@ -13380,23 +13011,6 @@ export const MODELS = {
13380
13011
  contextWindow: 400000,
13381
13012
  maxTokens: 128000,
13382
13013
  },
13383
- "openai/gpt-5.1-chat": {
13384
- id: "openai/gpt-5.1-chat",
13385
- name: "OpenAI: GPT-5.1 Chat",
13386
- api: "openai-completions",
13387
- provider: "openrouter",
13388
- baseUrl: "https://openrouter.ai/api/v1",
13389
- reasoning: false,
13390
- input: ["text", "image"],
13391
- cost: {
13392
- input: 1.25,
13393
- output: 10,
13394
- cacheRead: 0.13,
13395
- cacheWrite: 0,
13396
- },
13397
- contextWindow: 128000,
13398
- maxTokens: 32000,
13399
- },
13400
13014
  "openai/gpt-5.1-codex": {
13401
13015
  id: "openai/gpt-5.1-codex",
13402
13016
  name: "OpenAI: GPT-5.1-Codex",
@@ -13448,23 +13062,6 @@ export const MODELS = {
13448
13062
  contextWindow: 400000,
13449
13063
  maxTokens: 128000,
13450
13064
  },
13451
- "openai/gpt-5.1:batch": {
13452
- id: "openai/gpt-5.1:batch",
13453
- name: "OpenAI: GPT-5.1 (batch)",
13454
- api: "openai-completions",
13455
- provider: "openrouter",
13456
- baseUrl: "https://openrouter.ai/api/v1",
13457
- reasoning: true,
13458
- input: ["text", "image"],
13459
- cost: {
13460
- input: 0.625,
13461
- output: 5,
13462
- cacheRead: 0.0625,
13463
- cacheWrite: 0,
13464
- },
13465
- contextWindow: 400000,
13466
- maxTokens: 128000,
13467
- },
13468
13065
  "openai/gpt-5.2": {
13469
13066
  id: "openai/gpt-5.2",
13470
13067
  name: "OpenAI: GPT-5.2",
@@ -13537,24 +13134,6 @@ export const MODELS = {
13537
13134
  contextWindow: 400000,
13538
13135
  maxTokens: 128000,
13539
13136
  },
13540
- "openai/gpt-5.2:batch": {
13541
- id: "openai/gpt-5.2:batch",
13542
- name: "OpenAI: GPT-5.2 (batch)",
13543
- api: "openai-completions",
13544
- provider: "openrouter",
13545
- baseUrl: "https://openrouter.ai/api/v1",
13546
- reasoning: true,
13547
- thinkingLevelMap: { "xhigh": "xhigh" },
13548
- input: ["text", "image"],
13549
- cost: {
13550
- input: 0.875,
13551
- output: 7,
13552
- cacheRead: 0.0875,
13553
- cacheWrite: 0,
13554
- },
13555
- contextWindow: 400000,
13556
- maxTokens: 128000,
13557
- },
13558
13137
  "openai/gpt-5.3-chat": {
13559
13138
  id: "openai/gpt-5.3-chat",
13560
13139
  name: "OpenAI: GPT-5.3 Chat",
@@ -13627,24 +13206,6 @@ export const MODELS = {
13627
13206
  contextWindow: 400000,
13628
13207
  maxTokens: 128000,
13629
13208
  },
13630
- "openai/gpt-5.4-mini:batch": {
13631
- id: "openai/gpt-5.4-mini:batch",
13632
- name: "OpenAI: GPT-5.4 Mini (batch)",
13633
- api: "openai-completions",
13634
- provider: "openrouter",
13635
- baseUrl: "https://openrouter.ai/api/v1",
13636
- reasoning: true,
13637
- thinkingLevelMap: { "xhigh": "xhigh" },
13638
- input: ["text", "image"],
13639
- cost: {
13640
- input: 0.375,
13641
- output: 2.25,
13642
- cacheRead: 0.0375,
13643
- cacheWrite: 0,
13644
- },
13645
- contextWindow: 400000,
13646
- maxTokens: 128000,
13647
- },
13648
13209
  "openai/gpt-5.4-nano": {
13649
13210
  id: "openai/gpt-5.4-nano",
13650
13211
  name: "OpenAI: GPT-5.4 Nano",
@@ -13663,24 +13224,6 @@ export const MODELS = {
13663
13224
  contextWindow: 400000,
13664
13225
  maxTokens: 128000,
13665
13226
  },
13666
- "openai/gpt-5.4-nano:batch": {
13667
- id: "openai/gpt-5.4-nano:batch",
13668
- name: "OpenAI: GPT-5.4 Nano (batch)",
13669
- api: "openai-completions",
13670
- provider: "openrouter",
13671
- baseUrl: "https://openrouter.ai/api/v1",
13672
- reasoning: true,
13673
- thinkingLevelMap: { "xhigh": "xhigh" },
13674
- input: ["text", "image"],
13675
- cost: {
13676
- input: 0.09999999999999999,
13677
- output: 0.625,
13678
- cacheRead: 0.01,
13679
- cacheWrite: 0,
13680
- },
13681
- contextWindow: 400000,
13682
- maxTokens: 128000,
13683
- },
13684
13227
  "openai/gpt-5.4-pro": {
13685
13228
  id: "openai/gpt-5.4-pro",
13686
13229
  name: "OpenAI: GPT-5.4 Pro",
@@ -13699,24 +13242,6 @@ export const MODELS = {
13699
13242
  contextWindow: 1050000,
13700
13243
  maxTokens: 128000,
13701
13244
  },
13702
- "openai/gpt-5.4:batch": {
13703
- id: "openai/gpt-5.4:batch",
13704
- name: "OpenAI: GPT-5.4 (batch)",
13705
- api: "openai-completions",
13706
- provider: "openrouter",
13707
- baseUrl: "https://openrouter.ai/api/v1",
13708
- reasoning: true,
13709
- thinkingLevelMap: { "xhigh": "xhigh" },
13710
- input: ["text", "image"],
13711
- cost: {
13712
- input: 1.25,
13713
- output: 7.5,
13714
- cacheRead: 0.125,
13715
- cacheWrite: 0,
13716
- },
13717
- contextWindow: 1050000,
13718
- maxTokens: 128000,
13719
- },
13720
13245
  "openai/gpt-5.5": {
13721
13246
  id: "openai/gpt-5.5",
13722
13247
  name: "OpenAI: GPT-5.5",
@@ -13753,24 +13278,6 @@ export const MODELS = {
13753
13278
  contextWindow: 1050000,
13754
13279
  maxTokens: 128000,
13755
13280
  },
13756
- "openai/gpt-5.5:batch": {
13757
- id: "openai/gpt-5.5:batch",
13758
- name: "OpenAI: GPT-5.5 (batch)",
13759
- api: "openai-completions",
13760
- provider: "openrouter",
13761
- baseUrl: "https://openrouter.ai/api/v1",
13762
- reasoning: true,
13763
- thinkingLevelMap: { "xhigh": "xhigh" },
13764
- input: ["text", "image"],
13765
- cost: {
13766
- input: 2.5,
13767
- output: 15,
13768
- cacheRead: 0.25,
13769
- cacheWrite: 0,
13770
- },
13771
- contextWindow: 1050000,
13772
- maxTokens: 128000,
13773
- },
13774
13281
  "openai/gpt-5.6-luna": {
13775
13282
  id: "openai/gpt-5.6-luna",
13776
13283
  name: "OpenAI: GPT-5.6 Luna",
@@ -13879,23 +13386,6 @@ export const MODELS = {
13879
13386
  contextWindow: 1050000,
13880
13387
  maxTokens: 128000,
13881
13388
  },
13882
- "openai/gpt-5:batch": {
13883
- id: "openai/gpt-5:batch",
13884
- name: "OpenAI: GPT-5 (batch)",
13885
- api: "openai-completions",
13886
- provider: "openrouter",
13887
- baseUrl: "https://openrouter.ai/api/v1",
13888
- reasoning: true,
13889
- input: ["text", "image"],
13890
- cost: {
13891
- input: 0.625,
13892
- output: 5,
13893
- cacheRead: 0.0625,
13894
- cacheWrite: 0,
13895
- },
13896
- contextWindow: 400000,
13897
- maxTokens: 128000,
13898
- },
13899
13389
  "openai/gpt-audio": {
13900
13390
  id: "openai/gpt-audio",
13901
13391
  name: "OpenAI: GPT Audio",
@@ -13974,8 +13464,8 @@ export const MODELS = {
13974
13464
  input: ["text"],
13975
13465
  cost: {
13976
13466
  input: 0.03,
13977
- output: 0.13,
13978
- cacheRead: 0.03,
13467
+ output: 0.14,
13468
+ cacheRead: 0,
13979
13469
  cacheWrite: 0,
13980
13470
  },
13981
13471
  contextWindow: 131072,
@@ -14526,12 +14016,12 @@ export const MODELS = {
14526
14016
  input: ["text"],
14527
14017
  cost: {
14528
14018
  input: 0.07,
14529
- output: 0.27,
14019
+ output: 0.28,
14530
14020
  cacheRead: 0,
14531
14021
  cacheWrite: 0,
14532
14022
  },
14533
14023
  contextWindow: 262144,
14534
- maxTokens: 32768,
14024
+ maxTokens: 262144,
14535
14025
  },
14536
14026
  "qwen/qwen3-coder-flash": {
14537
14027
  id: "qwen/qwen3-coder-flash",
@@ -15234,6 +14724,23 @@ export const MODELS = {
15234
14724
  contextWindow: 1048576,
15235
14725
  maxTokens: 4096,
15236
14726
  },
14727
+ "thinkingmachines/inkling-small": {
14728
+ id: "thinkingmachines/inkling-small",
14729
+ name: "Thinking Machines: Inkling Small",
14730
+ api: "openai-completions",
14731
+ provider: "openrouter",
14732
+ baseUrl: "https://openrouter.ai/api/v1",
14733
+ reasoning: true,
14734
+ input: ["text", "image"],
14735
+ cost: {
14736
+ input: 0.5,
14737
+ output: 1.2,
14738
+ cacheRead: 0.09999999999999999,
14739
+ cacheWrite: 0,
14740
+ },
14741
+ contextWindow: 524288,
14742
+ maxTokens: 4096,
14743
+ },
15237
14744
  "upstage/solar-pro-3": {
15238
14745
  id: "upstage/solar-pro-3",
15239
14746
  name: "Upstage: Solar Pro 3",
@@ -15530,11 +15037,12 @@ export const MODELS = {
15530
15037
  provider: "openrouter",
15531
15038
  baseUrl: "https://openrouter.ai/api/v1",
15532
15039
  reasoning: true,
15040
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
15533
15041
  input: ["text"],
15534
15042
  cost: {
15535
- input: 1.232,
15536
- output: 3.8720000000000003,
15537
- cacheRead: 0.2288,
15043
+ input: 1.12,
15044
+ output: 3.52,
15045
+ cacheRead: 0.20800000000000002,
15538
15046
  cacheWrite: 0,
15539
15047
  },
15540
15048
  contextWindow: 1048576,
@@ -16038,7 +15546,7 @@ export const MODELS = {
16038
15546
  baseUrl: "https://api.together.ai/v1",
16039
15547
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
16040
15548
  reasoning: true,
16041
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
15549
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
16042
15550
  input: ["text"],
16043
15551
  cost: {
16044
15552
  input: 1.4,
@@ -17039,9 +16547,26 @@ export const MODELS = {
17039
16547
  reasoning: true,
17040
16548
  input: ["text"],
17041
16549
  cost: {
17042
- input: 0.13799999999999998,
17043
- output: 0.275,
17044
- cacheRead: 0.028,
16550
+ input: 0.14,
16551
+ output: 0.28,
16552
+ cacheRead: 0.0028,
16553
+ cacheWrite: 0,
16554
+ },
16555
+ contextWindow: 1000000,
16556
+ maxTokens: 384000,
16557
+ },
16558
+ "deepseek/deepseek-v4-flash-0731": {
16559
+ id: "deepseek/deepseek-v4-flash-0731",
16560
+ name: "DeepSeek V4 Flash 0731",
16561
+ api: "anthropic-messages",
16562
+ provider: "vercel-ai-gateway",
16563
+ baseUrl: "https://ai-gateway.vercel.sh",
16564
+ reasoning: true,
16565
+ input: ["text"],
16566
+ cost: {
16567
+ input: 0.13,
16568
+ output: 0.26,
16569
+ cacheRead: 0.0028,
17045
16570
  cacheWrite: 0,
17046
16571
  },
17047
16572
  contextWindow: 1000000,
@@ -19346,6 +18871,7 @@ export const MODELS = {
19346
18871
  provider: "vercel-ai-gateway",
19347
18872
  baseUrl: "https://ai-gateway.vercel.sh",
19348
18873
  reasoning: true,
18874
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
19349
18875
  input: ["text"],
19350
18876
  cost: {
19351
18877
  input: 1.1,
@@ -19363,6 +18889,7 @@ export const MODELS = {
19363
18889
  provider: "vercel-ai-gateway",
19364
18890
  baseUrl: "https://ai-gateway.vercel.sh",
19365
18891
  reasoning: true,
18892
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
19366
18893
  input: ["text"],
19367
18894
  cost: {
19368
18895
  input: 2.0999999999999996,
@@ -19916,24 +19443,6 @@ export const MODELS = {
19916
19443
  },
19917
19444
  },
19918
19445
  "zai": {
19919
- "glm-4.5-air": {
19920
- id: "glm-4.5-air",
19921
- name: "GLM-4.5-Air",
19922
- api: "openai-completions",
19923
- provider: "zai",
19924
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
19925
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
19926
- reasoning: true,
19927
- input: ["text"],
19928
- cost: {
19929
- input: 0,
19930
- output: 0,
19931
- cacheRead: 0,
19932
- cacheWrite: 0,
19933
- },
19934
- contextWindow: 131072,
19935
- maxTokens: 98304,
19936
- },
19937
19446
  "glm-4.7": {
19938
19447
  id: "glm-4.7",
19939
19448
  name: "GLM-4.7",
@@ -19970,32 +19479,15 @@ export const MODELS = {
19970
19479
  contextWindow: 200000,
19971
19480
  maxTokens: 131072,
19972
19481
  },
19973
- "glm-5.1": {
19974
- id: "glm-5.1",
19975
- name: "GLM-5.1",
19976
- api: "openai-completions",
19977
- provider: "zai",
19978
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
19979
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
19980
- reasoning: true,
19981
- input: ["text"],
19982
- cost: {
19983
- input: 0,
19984
- output: 0,
19985
- cacheRead: 0,
19986
- cacheWrite: 0,
19987
- },
19988
- contextWindow: 200000,
19989
- maxTokens: 131072,
19990
- },
19991
19482
  "glm-5.2": {
19992
19483
  id: "glm-5.2",
19993
19484
  name: "GLM-5.2",
19994
19485
  api: "openai-completions",
19995
19486
  provider: "zai",
19996
19487
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
19997
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
19488
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "supportsReasoningEffort": true, "zaiToolStream": true },
19998
19489
  reasoning: true,
19490
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
19999
19491
  input: ["text"],
20000
19492
  cost: {
20001
19493
  input: 0,
@@ -20006,44 +19498,27 @@ export const MODELS = {
20006
19498
  contextWindow: 1000000,
20007
19499
  maxTokens: 131072,
20008
19500
  },
20009
- "glm-5v-turbo": {
20010
- id: "glm-5v-turbo",
20011
- name: "GLM-5V-Turbo",
19501
+ "glm-5.2-highspeed[1m]": {
19502
+ id: "glm-5.2-highspeed[1m]",
19503
+ name: "GLM-5.2 Highspeed",
20012
19504
  api: "openai-completions",
20013
19505
  provider: "zai",
20014
19506
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
20015
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
19507
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "supportsReasoningEffort": true, "zaiToolStream": true },
20016
19508
  reasoning: true,
20017
- input: ["text", "image"],
19509
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
19510
+ input: ["text"],
20018
19511
  cost: {
20019
19512
  input: 0,
20020
19513
  output: 0,
20021
19514
  cacheRead: 0,
20022
19515
  cacheWrite: 0,
20023
19516
  },
20024
- contextWindow: 200000,
19517
+ contextWindow: 1000000,
20025
19518
  maxTokens: 131072,
20026
19519
  },
20027
19520
  },
20028
19521
  "zai-coding-cn": {
20029
- "glm-4.5-air": {
20030
- id: "glm-4.5-air",
20031
- name: "GLM-4.5-Air",
20032
- api: "openai-completions",
20033
- provider: "zai-coding-cn",
20034
- baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
20035
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
20036
- reasoning: true,
20037
- input: ["text"],
20038
- cost: {
20039
- input: 0,
20040
- output: 0,
20041
- cacheRead: 0,
20042
- cacheWrite: 0,
20043
- },
20044
- contextWindow: 131072,
20045
- maxTokens: 98304,
20046
- },
20047
19522
  "glm-4.7": {
20048
19523
  id: "glm-4.7",
20049
19524
  name: "GLM-4.7",
@@ -20080,32 +19555,15 @@ export const MODELS = {
20080
19555
  contextWindow: 200000,
20081
19556
  maxTokens: 131072,
20082
19557
  },
20083
- "glm-5.1": {
20084
- id: "glm-5.1",
20085
- name: "GLM-5.1",
20086
- api: "openai-completions",
20087
- provider: "zai-coding-cn",
20088
- baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
20089
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
20090
- reasoning: true,
20091
- input: ["text"],
20092
- cost: {
20093
- input: 0,
20094
- output: 0,
20095
- cacheRead: 0,
20096
- cacheWrite: 0,
20097
- },
20098
- contextWindow: 200000,
20099
- maxTokens: 131072,
20100
- },
20101
19558
  "glm-5.2": {
20102
19559
  id: "glm-5.2",
20103
19560
  name: "GLM-5.2",
20104
19561
  api: "openai-completions",
20105
19562
  provider: "zai-coding-cn",
20106
19563
  baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
20107
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
19564
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "supportsReasoningEffort": true, "zaiToolStream": true },
20108
19565
  reasoning: true,
19566
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
20109
19567
  input: ["text"],
20110
19568
  cost: {
20111
19569
  input: 0,
@@ -20116,22 +19574,23 @@ export const MODELS = {
20116
19574
  contextWindow: 1000000,
20117
19575
  maxTokens: 131072,
20118
19576
  },
20119
- "glm-5v-turbo": {
20120
- id: "glm-5v-turbo",
20121
- name: "GLM-5V-Turbo",
19577
+ "glm-5.2-highspeed[1m]": {
19578
+ id: "glm-5.2-highspeed[1m]",
19579
+ name: "GLM-5.2 Highspeed",
20122
19580
  api: "openai-completions",
20123
19581
  provider: "zai-coding-cn",
20124
19582
  baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
20125
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
19583
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "supportsReasoningEffort": true, "zaiToolStream": true },
20126
19584
  reasoning: true,
20127
- input: ["text", "image"],
19585
+ thinkingLevelMap: { "off": null, "minimal": null, "low": "low", "medium": "medium", "high": "high", "xhigh": "xhigh", "max": "max" },
19586
+ input: ["text"],
20128
19587
  cost: {
20129
19588
  input: 0,
20130
19589
  output: 0,
20131
19590
  cacheRead: 0,
20132
19591
  cacheWrite: 0,
20133
19592
  },
20134
- contextWindow: 200000,
19593
+ contextWindow: 1000000,
20135
19594
  maxTokens: 131072,
20136
19595
  },
20137
19596
  },