@caupulican/pi-ai 0.85.3 → 0.85.4

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 (36) hide show
  1. package/dist/bedrock-provider.d.ts +1 -0
  2. package/dist/bedrock-provider.d.ts.map +1 -1
  3. package/dist/bedrock-provider.js +1 -0
  4. package/dist/bedrock-provider.js.map +1 -1
  5. package/dist/bedrock-scope.d.ts +35 -0
  6. package/dist/bedrock-scope.d.ts.map +1 -0
  7. package/dist/bedrock-scope.js +141 -0
  8. package/dist/bedrock-scope.js.map +1 -0
  9. package/dist/env-api-keys.d.ts.map +1 -1
  10. package/dist/env-api-keys.js +4 -1
  11. package/dist/env-api-keys.js.map +1 -1
  12. package/dist/image-models.generated.js +1 -1
  13. package/dist/image-models.generated.js.map +1 -1
  14. package/dist/models.generated.d.ts +217 -30
  15. package/dist/models.generated.d.ts.map +1 -1
  16. package/dist/models.generated.js +225 -80
  17. package/dist/models.generated.js.map +1 -1
  18. package/dist/stream.d.ts.map +1 -1
  19. package/dist/stream.js +193 -87
  20. package/dist/stream.js.map +1 -1
  21. package/dist/utils/tool-repair/registry.d.ts +21 -0
  22. package/dist/utils/tool-repair/registry.d.ts.map +1 -1
  23. package/dist/utils/tool-repair/registry.js +27 -0
  24. package/dist/utils/tool-repair/registry.js.map +1 -1
  25. package/dist/utils/tool-repair/text-protocol-history.d.ts.map +1 -1
  26. package/dist/utils/tool-repair/text-protocol-history.js +4 -1
  27. package/dist/utils/tool-repair/text-protocol-history.js.map +1 -1
  28. package/dist/utils/tool-repair/text-protocol-live-filter.d.ts +5 -0
  29. package/dist/utils/tool-repair/text-protocol-live-filter.d.ts.map +1 -1
  30. package/dist/utils/tool-repair/text-protocol-live-filter.js +11 -0
  31. package/dist/utils/tool-repair/text-protocol-live-filter.js.map +1 -1
  32. package/dist/utils/tool-repair/text-protocol.d.ts +5 -1
  33. package/dist/utils/tool-repair/text-protocol.d.ts.map +1 -1
  34. package/dist/utils/tool-repair/text-protocol.js +49 -27
  35. package/dist/utils/tool-repair/text-protocol.js.map +1 -1
  36. package/package.json +3 -1
@@ -4639,44 +4639,6 @@ export const MODELS = {
4639
4639
  contextWindow: 1000000,
4640
4640
  maxTokens: 128000,
4641
4641
  },
4642
- "gemini-2.5-pro": {
4643
- id: "gemini-2.5-pro",
4644
- name: "Gemini 2.5 Pro",
4645
- api: "openai-completions",
4646
- provider: "github-copilot",
4647
- baseUrl: "https://api.individual.githubcopilot.com",
4648
- headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4649
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4650
- reasoning: true,
4651
- input: ["text", "image"],
4652
- cost: {
4653
- input: 1.25,
4654
- output: 10,
4655
- cacheRead: 0.125,
4656
- cacheWrite: 0,
4657
- },
4658
- contextWindow: 128000,
4659
- maxTokens: 64000,
4660
- },
4661
- "gemini-3-flash-preview": {
4662
- id: "gemini-3-flash-preview",
4663
- name: "Gemini 3 Flash Preview",
4664
- api: "openai-completions",
4665
- provider: "github-copilot",
4666
- baseUrl: "https://api.individual.githubcopilot.com",
4667
- headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4668
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4669
- reasoning: true,
4670
- input: ["text", "image"],
4671
- cost: {
4672
- input: 0.5,
4673
- output: 3,
4674
- cacheRead: 0.05,
4675
- cacheWrite: 0,
4676
- },
4677
- contextWindow: 128000,
4678
- maxTokens: 64000,
4679
- },
4680
4642
  "gemini-3.1-pro-preview": {
4681
4643
  id: "gemini-3.1-pro-preview",
4682
4644
  name: "Gemini 3.1 Pro Preview",
@@ -4715,6 +4677,25 @@ export const MODELS = {
4715
4677
  contextWindow: 200000,
4716
4678
  maxTokens: 64000,
4717
4679
  },
4680
+ "gemini-3.6-flash": {
4681
+ id: "gemini-3.6-flash",
4682
+ name: "Gemini 3.6 Flash",
4683
+ api: "openai-completions",
4684
+ provider: "github-copilot",
4685
+ baseUrl: "https://api.individual.githubcopilot.com",
4686
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4687
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4688
+ reasoning: true,
4689
+ input: ["text", "image"],
4690
+ cost: {
4691
+ input: 1.5,
4692
+ output: 7.5,
4693
+ cacheRead: 0.15,
4694
+ cacheWrite: 0,
4695
+ },
4696
+ contextWindow: 1000000,
4697
+ maxTokens: 64000,
4698
+ },
4718
4699
  "gpt-4.1": {
4719
4700
  id: "gpt-4.1",
4720
4701
  name: "GPT-4.1",
@@ -4897,10 +4878,10 @@ export const MODELS = {
4897
4878
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh", "max": "max" },
4898
4879
  input: ["text", "image"],
4899
4880
  cost: {
4900
- input: 1,
4901
- output: 6,
4902
- cacheRead: 0.1,
4903
- cacheWrite: 1.25,
4881
+ input: 0.2,
4882
+ output: 1.2,
4883
+ cacheRead: 0.02,
4884
+ cacheWrite: 0,
4904
4885
  },
4905
4886
  contextWindow: 1050000,
4906
4887
  maxTokens: 128000,
@@ -4935,14 +4916,33 @@ export const MODELS = {
4935
4916
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh", "max": "max" },
4936
4917
  input: ["text", "image"],
4937
4918
  cost: {
4938
- input: 2.5,
4939
- output: 15,
4940
- cacheRead: 0.25,
4941
- cacheWrite: 3.125,
4919
+ input: 2,
4920
+ output: 12,
4921
+ cacheRead: 0.2,
4922
+ cacheWrite: 0,
4942
4923
  },
4943
4924
  contextWindow: 1050000,
4944
4925
  maxTokens: 128000,
4945
4926
  },
4927
+ "grok-4.5": {
4928
+ id: "grok-4.5",
4929
+ name: "Grok 4.5",
4930
+ api: "openai-completions",
4931
+ provider: "github-copilot",
4932
+ baseUrl: "https://api.individual.githubcopilot.com",
4933
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4934
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4935
+ reasoning: true,
4936
+ input: ["text", "image"],
4937
+ cost: {
4938
+ input: 2,
4939
+ output: 6,
4940
+ cacheRead: 0.5,
4941
+ cacheWrite: 0,
4942
+ },
4943
+ contextWindow: 500000,
4944
+ maxTokens: 128000,
4945
+ },
4946
4946
  "kimi-k2.7-code": {
4947
4947
  id: "kimi-k2.7-code",
4948
4948
  name: "Kimi K2.7 Code",
@@ -6240,6 +6240,24 @@ export const MODELS = {
6240
6240
  contextWindow: 1048576,
6241
6241
  maxTokens: 384000,
6242
6242
  },
6243
+ "deepseek-ai/DeepSeek-V4-Flash-0731": {
6244
+ id: "deepseek-ai/DeepSeek-V4-Flash-0731",
6245
+ name: "DeepSeek V4 Flash 0731",
6246
+ api: "openai-completions",
6247
+ provider: "huggingface",
6248
+ baseUrl: "https://router.huggingface.co/v1",
6249
+ compat: { "supportsDeveloperRole": false },
6250
+ reasoning: true,
6251
+ input: ["text"],
6252
+ cost: {
6253
+ input: 0.14,
6254
+ output: 0.28,
6255
+ cacheRead: 0,
6256
+ cacheWrite: 0,
6257
+ },
6258
+ contextWindow: 1048576,
6259
+ maxTokens: 384000,
6260
+ },
6243
6261
  "deepseek-ai/DeepSeek-V4-Pro": {
6244
6262
  id: "deepseek-ai/DeepSeek-V4-Pro",
6245
6263
  name: "DeepSeek V4 Pro",
@@ -10058,6 +10076,23 @@ export const MODELS = {
10058
10076
  contextWindow: 1000000,
10059
10077
  maxTokens: 65536,
10060
10078
  },
10079
+ "qwen3.8-max": {
10080
+ id: "qwen3.8-max",
10081
+ name: "Qwen3.8 Max",
10082
+ api: "anthropic-messages",
10083
+ provider: "opencode-go",
10084
+ baseUrl: "https://opencode.ai/zen/go",
10085
+ reasoning: true,
10086
+ input: ["text", "image"],
10087
+ cost: {
10088
+ input: 2,
10089
+ output: 6,
10090
+ cacheRead: 0.25,
10091
+ cacheWrite: 2.5,
10092
+ },
10093
+ contextWindow: 1000000,
10094
+ maxTokens: 131072,
10095
+ },
10061
10096
  },
10062
10097
  "openrouter": {
10063
10098
  "ai21/jamba-large-1.7": {
@@ -10837,7 +10872,7 @@ export const MODELS = {
10837
10872
  },
10838
10873
  "deepseek/deepseek-v4-flash": {
10839
10874
  id: "deepseek/deepseek-v4-flash",
10840
- name: "DeepSeek: DeepSeek V4 Flash",
10875
+ name: "DeepSeek: DeepSeek V4 Flash 0423",
10841
10876
  api: "openai-completions",
10842
10877
  provider: "openrouter",
10843
10878
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10904,7 +10939,7 @@ export const MODELS = {
10904
10939
  input: 0.3,
10905
10940
  output: 2.5,
10906
10941
  cacheRead: 0.03,
10907
- cacheWrite: 0.08333333333333334,
10942
+ cacheWrite: 0.0833333333333333,
10908
10943
  },
10909
10944
  contextWindow: 1048576,
10910
10945
  maxTokens: 65535,
@@ -10921,7 +10956,7 @@ export const MODELS = {
10921
10956
  input: 0.09999999999999999,
10922
10957
  output: 0.39999999999999997,
10923
10958
  cacheRead: 0.01,
10924
- cacheWrite: 0.08333333333333334,
10959
+ cacheWrite: 0.0833333333333333,
10925
10960
  },
10926
10961
  contextWindow: 1048576,
10927
10962
  maxTokens: 65535,
@@ -10989,7 +11024,7 @@ export const MODELS = {
10989
11024
  input: 0.5,
10990
11025
  output: 3,
10991
11026
  cacheRead: 0.049999999999999996,
10992
- cacheWrite: 0.08333333333333334,
11027
+ cacheWrite: 0.0833333333333333,
10993
11028
  },
10994
11029
  contextWindow: 1048576,
10995
11030
  maxTokens: 65535,
@@ -11023,7 +11058,7 @@ export const MODELS = {
11023
11058
  input: 0.25,
11024
11059
  output: 1.5,
11025
11060
  cacheRead: 0.024999999999999998,
11026
- cacheWrite: 0.08333333333333334,
11061
+ cacheWrite: 0.0833333333333333,
11027
11062
  },
11028
11063
  contextWindow: 1048576,
11029
11064
  maxTokens: 65536,
@@ -11040,7 +11075,7 @@ export const MODELS = {
11040
11075
  input: 0.25,
11041
11076
  output: 1.5,
11042
11077
  cacheRead: 0.024999999999999998,
11043
- cacheWrite: 0.08333333333333334,
11078
+ cacheWrite: 0.0833333333333333,
11044
11079
  },
11045
11080
  contextWindow: 1048576,
11046
11081
  maxTokens: 65536,
@@ -11091,7 +11126,7 @@ export const MODELS = {
11091
11126
  input: 1.5,
11092
11127
  output: 9,
11093
11128
  cacheRead: 0.15,
11094
- cacheWrite: 0.08333333333333334,
11129
+ cacheWrite: 0.0833333333333333,
11095
11130
  },
11096
11131
  contextWindow: 1048576,
11097
11132
  maxTokens: 65536,
@@ -11108,7 +11143,7 @@ export const MODELS = {
11108
11143
  input: 0.3,
11109
11144
  output: 2.5,
11110
11145
  cacheRead: 0.03,
11111
- cacheWrite: 0.08333333333333334,
11146
+ cacheWrite: 0.0833333333333333,
11112
11147
  },
11113
11148
  contextWindow: 1048576,
11114
11149
  maxTokens: 65536,
@@ -11125,7 +11160,7 @@ export const MODELS = {
11125
11160
  input: 1.5,
11126
11161
  output: 7.5,
11127
11162
  cacheRead: 0.15,
11128
- cacheWrite: 0.08333333333333334,
11163
+ cacheWrite: 0.0833333333333333,
11129
11164
  },
11130
11165
  contextWindow: 1048576,
11131
11166
  maxTokens: 65536,
@@ -11446,13 +11481,13 @@ export const MODELS = {
11446
11481
  reasoning: false,
11447
11482
  input: ["text"],
11448
11483
  cost: {
11449
- input: 0.13,
11450
- output: 0.39999999999999997,
11484
+ input: 0.09999999999999999,
11485
+ output: 0.32,
11451
11486
  cacheRead: 0,
11452
11487
  cacheWrite: 0,
11453
11488
  },
11454
11489
  contextWindow: 131072,
11455
- maxTokens: 128000,
11490
+ maxTokens: 16384,
11456
11491
  },
11457
11492
  "meta-llama/llama-4-maverick": {
11458
11493
  id: "meta-llama/llama-4-maverick",
@@ -11957,9 +11992,9 @@ export const MODELS = {
11957
11992
  reasoning: true,
11958
11993
  input: ["text", "image"],
11959
11994
  cost: {
11960
- input: 0.6,
11961
- output: 3.41,
11962
- cacheRead: 0.19999999999999998,
11995
+ input: 0.589,
11996
+ output: 2.48,
11997
+ cacheRead: 0.0992,
11963
11998
  cacheWrite: 0,
11964
11999
  },
11965
12000
  contextWindow: 262144,
@@ -13540,12 +13575,12 @@ export const MODELS = {
13540
13575
  input: ["text"],
13541
13576
  cost: {
13542
13577
  input: 0.07,
13543
- output: 0.28,
13578
+ output: 0.27,
13544
13579
  cacheRead: 0,
13545
13580
  cacheWrite: 0,
13546
13581
  },
13547
13582
  contextWindow: 262144,
13548
- maxTokens: 262144,
13583
+ maxTokens: 32768,
13549
13584
  },
13550
13585
  "qwen/qwen3-coder-flash": {
13551
13586
  id: "qwen/qwen3-coder-flash",
@@ -13641,13 +13676,13 @@ export const MODELS = {
13641
13676
  reasoning: false,
13642
13677
  input: ["text"],
13643
13678
  cost: {
13644
- input: 0.09999999999999999,
13679
+ input: 0.09,
13645
13680
  output: 1.1,
13646
- cacheRead: 0.07,
13681
+ cacheRead: 0,
13647
13682
  cacheWrite: 0,
13648
13683
  },
13649
13684
  contextWindow: 262144,
13650
- maxTokens: 262144,
13685
+ maxTokens: 16384,
13651
13686
  },
13652
13687
  "qwen/qwen3-next-80b-a3b-thinking": {
13653
13688
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -13664,7 +13699,7 @@ export const MODELS = {
13664
13699
  cacheWrite: 0,
13665
13700
  },
13666
13701
  contextWindow: 262144,
13667
- maxTokens: 32768,
13702
+ maxTokens: 4096,
13668
13703
  },
13669
13704
  "qwen/qwen3-vl-235b-a22b-instruct": {
13670
13705
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -13709,13 +13744,13 @@ export const MODELS = {
13709
13744
  reasoning: false,
13710
13745
  input: ["text", "image"],
13711
13746
  cost: {
13712
- input: 0.13,
13713
- output: 0.52,
13747
+ input: 0.15,
13748
+ output: 0.6,
13714
13749
  cacheRead: 0,
13715
13750
  cacheWrite: 0,
13716
13751
  },
13717
13752
  contextWindow: 262144,
13718
- maxTokens: 32768,
13753
+ maxTokens: 16384,
13719
13754
  },
13720
13755
  "qwen/qwen3-vl-30b-a3b-thinking": {
13721
13756
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -13794,8 +13829,8 @@ export const MODELS = {
13794
13829
  reasoning: true,
13795
13830
  input: ["text", "image"],
13796
13831
  cost: {
13797
- input: 0.39999999999999997,
13798
- output: 3.1999999999999997,
13832
+ input: 0.26,
13833
+ output: 2.08,
13799
13834
  cacheRead: 0,
13800
13835
  cacheWrite: 0,
13801
13836
  },
@@ -13930,13 +13965,13 @@ export const MODELS = {
13930
13965
  reasoning: true,
13931
13966
  input: ["text", "image"],
13932
13967
  cost: {
13933
- input: 0.3,
13934
- output: 2,
13935
- cacheRead: 0.15,
13968
+ input: 0.28900000000000003,
13969
+ output: 2.4,
13970
+ cacheRead: 0,
13936
13971
  cacheWrite: 0,
13937
13972
  },
13938
13973
  contextWindow: 262144,
13939
- maxTokens: 65536,
13974
+ maxTokens: 131072,
13940
13975
  },
13941
13976
  "qwen/qwen3.6-35b-a3b": {
13942
13977
  id: "qwen/qwen3.6-35b-a3b",
@@ -14057,6 +14092,23 @@ export const MODELS = {
14057
14092
  contextWindow: 1000000,
14058
14093
  maxTokens: 131072,
14059
14094
  },
14095
+ "qwen/qwen3.8-max": {
14096
+ id: "qwen/qwen3.8-max",
14097
+ name: "Qwen: Qwen3.8 Max",
14098
+ api: "openai-completions",
14099
+ provider: "openrouter",
14100
+ baseUrl: "https://openrouter.ai/api/v1",
14101
+ reasoning: true,
14102
+ input: ["text", "image"],
14103
+ cost: {
14104
+ input: 2,
14105
+ output: 6,
14106
+ cacheRead: 0.25,
14107
+ cacheWrite: 2.5,
14108
+ },
14109
+ contextWindow: 1000000,
14110
+ maxTokens: 131072,
14111
+ },
14060
14112
  "rekaai/reka-edge": {
14061
14113
  id: "rekaai/reka-edge",
14062
14114
  name: "Reka Edge",
@@ -14542,9 +14594,9 @@ export const MODELS = {
14542
14594
  reasoning: true,
14543
14595
  input: ["text"],
14544
14596
  cost: {
14545
- input: 1.19,
14546
- output: 3.74,
14547
- cacheRead: 0.221,
14597
+ input: 0.76,
14598
+ output: 2.42,
14599
+ cacheRead: 0.14,
14548
14600
  cacheWrite: 0,
14549
14601
  },
14550
14602
  contextWindow: 1048576,
@@ -14670,7 +14722,7 @@ export const MODELS = {
14670
14722
  input: 1.5,
14671
14723
  output: 7.5,
14672
14724
  cacheRead: 0.15,
14673
- cacheWrite: 0.08333333333333334,
14725
+ cacheWrite: 0.0833333333333333,
14674
14726
  },
14675
14727
  contextWindow: 1048576,
14676
14728
  maxTokens: 65536,
@@ -14817,6 +14869,25 @@ export const MODELS = {
14817
14869
  contextWindow: 1000000,
14818
14870
  maxTokens: 384000,
14819
14871
  },
14872
+ "deepseek-v4-flash-0731": {
14873
+ id: "deepseek-v4-flash-0731",
14874
+ name: "DeepSeek V4 Flash 0731",
14875
+ api: "openai-completions",
14876
+ provider: "qwen-token-plan",
14877
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14878
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14879
+ reasoning: true,
14880
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
14881
+ input: ["text"],
14882
+ cost: {
14883
+ input: 0,
14884
+ output: 0,
14885
+ cacheRead: 0,
14886
+ cacheWrite: 0,
14887
+ },
14888
+ contextWindow: 1000000,
14889
+ maxTokens: 384000,
14890
+ },
14820
14891
  "deepseek-v4-pro": {
14821
14892
  id: "deepseek-v4-pro",
14822
14893
  name: "DeepSeek V4 Pro",
@@ -15016,6 +15087,24 @@ export const MODELS = {
15016
15087
  contextWindow: 1000000,
15017
15088
  maxTokens: 65536,
15018
15089
  },
15090
+ "qwen3.8-max": {
15091
+ id: "qwen3.8-max",
15092
+ name: "Qwen3.8 Max",
15093
+ api: "openai-completions",
15094
+ provider: "qwen-token-plan",
15095
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
15096
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15097
+ reasoning: true,
15098
+ input: ["text", "image"],
15099
+ cost: {
15100
+ input: 0,
15101
+ output: 0,
15102
+ cacheRead: 0,
15103
+ cacheWrite: 0,
15104
+ },
15105
+ contextWindow: 1000000,
15106
+ maxTokens: 131072,
15107
+ },
15019
15108
  "qwen3.8-max-preview": {
15020
15109
  id: "qwen3.8-max-preview",
15021
15110
  name: "Qwen3.8 Max Preview",
@@ -15091,6 +15180,25 @@ export const MODELS = {
15091
15180
  contextWindow: 1000000,
15092
15181
  maxTokens: 384000,
15093
15182
  },
15183
+ "deepseek-v4-flash-0731": {
15184
+ id: "deepseek-v4-flash-0731",
15185
+ name: "DeepSeek V4 Flash 0731",
15186
+ api: "openai-completions",
15187
+ provider: "qwen-token-plan-cn",
15188
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15189
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15190
+ reasoning: true,
15191
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
15192
+ input: ["text"],
15193
+ cost: {
15194
+ input: 0,
15195
+ output: 0,
15196
+ cacheRead: 0,
15197
+ cacheWrite: 0,
15198
+ },
15199
+ contextWindow: 1000000,
15200
+ maxTokens: 384000,
15201
+ },
15094
15202
  "deepseek-v4-pro": {
15095
15203
  id: "deepseek-v4-pro",
15096
15204
  name: "DeepSeek V4 Pro",
@@ -15290,6 +15398,24 @@ export const MODELS = {
15290
15398
  contextWindow: 1000000,
15291
15399
  maxTokens: 65536,
15292
15400
  },
15401
+ "qwen3.8-max": {
15402
+ id: "qwen3.8-max",
15403
+ name: "Qwen3.8 Max",
15404
+ api: "openai-completions",
15405
+ provider: "qwen-token-plan-cn",
15406
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15407
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15408
+ reasoning: true,
15409
+ input: ["text", "image"],
15410
+ cost: {
15411
+ input: 0,
15412
+ output: 0,
15413
+ cacheRead: 0,
15414
+ cacheWrite: 0,
15415
+ },
15416
+ contextWindow: 1000000,
15417
+ maxTokens: 131072,
15418
+ },
15293
15419
  "qwen3.8-max-preview": {
15294
15420
  id: "qwen3.8-max-preview",
15295
15421
  name: "Qwen3.8 Max Preview",
@@ -15422,6 +15548,25 @@ export const MODELS = {
15422
15548
  contextWindow: 1000000,
15423
15549
  maxTokens: 500000,
15424
15550
  },
15551
+ "deepseek-ai/DeepSeek-V4-Flash-0731": {
15552
+ id: "deepseek-ai/DeepSeek-V4-Flash-0731",
15553
+ name: "DeepSeek V4 Flash 0731",
15554
+ api: "openai-completions",
15555
+ provider: "together",
15556
+ baseUrl: "https://api.together.ai/v1",
15557
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
15558
+ reasoning: true,
15559
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
15560
+ input: ["text"],
15561
+ cost: {
15562
+ input: 0.14,
15563
+ output: 0.28,
15564
+ cacheRead: 0.03,
15565
+ cacheWrite: 0,
15566
+ },
15567
+ contextWindow: 1000000,
15568
+ maxTokens: 384000,
15569
+ },
15425
15570
  "deepseek-ai/DeepSeek-V4-Pro": {
15426
15571
  id: "deepseek-ai/DeepSeek-V4-Pro",
15427
15572
  name: "DeepSeek V4 Pro",