@code-yeongyu/senpi 2026.6.10-2 → 2026.6.12-2

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 (82) hide show
  1. package/CHANGELOG.md +22 -0
  2. package/dist/cli/startup-ui.d.ts +10 -0
  3. package/dist/cli/startup-ui.d.ts.map +1 -1
  4. package/dist/cli/startup-ui.js +72 -1
  5. package/dist/cli/startup-ui.js.map +1 -1
  6. package/dist/core/bash-executor.d.ts.map +1 -1
  7. package/dist/core/bash-executor.js +31 -7
  8. package/dist/core/bash-executor.js.map +1 -1
  9. package/dist/core/extensions/types.d.ts +1 -0
  10. package/dist/core/extensions/types.d.ts.map +1 -1
  11. package/dist/core/extensions/types.js.map +1 -1
  12. package/dist/core/model-resolver.d.ts +1 -0
  13. package/dist/core/model-resolver.d.ts.map +1 -1
  14. package/dist/core/model-resolver.js +20 -5
  15. package/dist/core/model-resolver.js.map +1 -1
  16. package/dist/core/settings-manager.d.ts +6 -0
  17. package/dist/core/settings-manager.d.ts.map +1 -1
  18. package/dist/core/settings-manager.js +17 -0
  19. package/dist/core/settings-manager.js.map +1 -1
  20. package/dist/core/tools/output-accumulator.d.ts +3 -5
  21. package/dist/core/tools/output-accumulator.d.ts.map +1 -1
  22. package/dist/core/tools/output-accumulator.js +38 -34
  23. package/dist/core/tools/output-accumulator.js.map +1 -1
  24. package/dist/core/tools/tail-window.d.ts +15 -0
  25. package/dist/core/tools/tail-window.d.ts.map +1 -0
  26. package/dist/core/tools/tail-window.js +75 -0
  27. package/dist/core/tools/tail-window.js.map +1 -0
  28. package/dist/main.d.ts.map +1 -1
  29. package/dist/main.js +8 -1
  30. package/dist/main.js.map +1 -1
  31. package/dist/modes/interactive/components/first-time-setup.d.ts +25 -0
  32. package/dist/modes/interactive/components/first-time-setup.d.ts.map +1 -0
  33. package/dist/modes/interactive/components/first-time-setup.js +101 -0
  34. package/dist/modes/interactive/components/first-time-setup.js.map +1 -0
  35. package/dist/modes/interactive/components/index.d.ts +1 -0
  36. package/dist/modes/interactive/components/index.d.ts.map +1 -1
  37. package/dist/modes/interactive/components/index.js +1 -0
  38. package/dist/modes/interactive/components/index.js.map +1 -1
  39. package/dist/modes/interactive/components/tool-execution.d.ts +6 -0
  40. package/dist/modes/interactive/components/tool-execution.d.ts.map +1 -1
  41. package/dist/modes/interactive/components/tool-execution.js +45 -0
  42. package/dist/modes/interactive/components/tool-execution.js.map +1 -1
  43. package/dist/modes/interactive/interactive-mode.d.ts +1 -0
  44. package/dist/modes/interactive/interactive-mode.d.ts.map +1 -1
  45. package/dist/modes/interactive/interactive-mode.js +12 -5
  46. package/dist/modes/interactive/interactive-mode.js.map +1 -1
  47. package/docs/settings.md +2 -0
  48. package/node_modules/@earendil-works/pi-agent-core/package.json +2 -2
  49. package/node_modules/@earendil-works/pi-ai/dist/models.d.ts.map +1 -1
  50. package/node_modules/@earendil-works/pi-ai/dist/models.generated.d.ts +92 -227
  51. package/node_modules/@earendil-works/pi-ai/dist/models.generated.d.ts.map +1 -1
  52. package/node_modules/@earendil-works/pi-ai/dist/models.generated.js +140 -247
  53. package/node_modules/@earendil-works/pi-ai/dist/models.generated.js.map +1 -1
  54. package/node_modules/@earendil-works/pi-ai/dist/models.js +12 -5
  55. package/node_modules/@earendil-works/pi-ai/dist/models.js.map +1 -1
  56. package/node_modules/@earendil-works/pi-ai/dist/providers/amazon-bedrock.d.ts.map +1 -1
  57. package/node_modules/@earendil-works/pi-ai/dist/providers/amazon-bedrock.js +11 -2
  58. package/node_modules/@earendil-works/pi-ai/dist/providers/amazon-bedrock.js.map +1 -1
  59. package/node_modules/@earendil-works/pi-ai/dist/utils/event-stream.d.ts +4 -1
  60. package/node_modules/@earendil-works/pi-ai/dist/utils/event-stream.d.ts.map +1 -1
  61. package/node_modules/@earendil-works/pi-ai/dist/utils/event-stream.js +54 -8
  62. package/node_modules/@earendil-works/pi-ai/dist/utils/event-stream.js.map +1 -1
  63. package/node_modules/@earendil-works/pi-ai/package.json +1 -1
  64. package/node_modules/@earendil-works/pi-tui/dist/components/editor.d.ts +2 -0
  65. package/node_modules/@earendil-works/pi-tui/dist/components/editor.d.ts.map +1 -1
  66. package/node_modules/@earendil-works/pi-tui/dist/components/editor.js +80 -7
  67. package/node_modules/@earendil-works/pi-tui/dist/components/editor.js.map +1 -1
  68. package/node_modules/@earendil-works/pi-tui/dist/components/markdown.d.ts +1 -0
  69. package/node_modules/@earendil-works/pi-tui/dist/components/markdown.d.ts.map +1 -1
  70. package/node_modules/@earendil-works/pi-tui/dist/components/markdown.js +72 -1
  71. package/node_modules/@earendil-works/pi-tui/dist/components/markdown.js.map +1 -1
  72. package/node_modules/@earendil-works/pi-tui/dist/index.d.ts +1 -1
  73. package/node_modules/@earendil-works/pi-tui/dist/index.d.ts.map +1 -1
  74. package/node_modules/@earendil-works/pi-tui/dist/index.js +1 -1
  75. package/node_modules/@earendil-works/pi-tui/dist/index.js.map +1 -1
  76. package/node_modules/@earendil-works/pi-tui/dist/utils.d.ts +1 -0
  77. package/node_modules/@earendil-works/pi-tui/dist/utils.d.ts.map +1 -1
  78. package/node_modules/@earendil-works/pi-tui/dist/utils.js +1 -1
  79. package/node_modules/@earendil-works/pi-tui/dist/utils.js.map +1 -1
  80. package/node_modules/@earendil-works/pi-tui/package.json +1 -1
  81. package/npm-shrinkwrap.json +12 -12
  82. package/package.json +4 -4
@@ -2988,30 +2988,13 @@ export const MODELS = {
2988
2988
  reasoning: true,
2989
2989
  input: ["text"],
2990
2990
  cost: {
2991
- input: 0.25,
2992
- output: 0.69,
2991
+ input: 0.35,
2992
+ output: 0.75,
2993
2993
  cacheRead: 0,
2994
2994
  cacheWrite: 0,
2995
2995
  },
2996
2996
  contextWindow: 131072,
2997
- maxTokens: 32768,
2998
- },
2999
- "llama3.1-8b": {
3000
- id: "llama3.1-8b",
3001
- name: "Llama 3.1 8B",
3002
- api: "openai-completions",
3003
- provider: "cerebras",
3004
- baseUrl: "https://api.cerebras.ai/v1",
3005
- reasoning: false,
3006
- input: ["text"],
3007
- cost: {
3008
- input: 0.1,
3009
- output: 0.1,
3010
- cacheRead: 0,
3011
- cacheWrite: 0,
3012
- },
3013
- contextWindow: 32000,
3014
- maxTokens: 8000,
2997
+ maxTokens: 40960,
3015
2998
  },
3016
2999
  "zai-glm-4.7": {
3017
3000
  id: "zai-glm-4.7",
@@ -3019,7 +3002,7 @@ export const MODELS = {
3019
3002
  api: "openai-completions",
3020
3003
  provider: "cerebras",
3021
3004
  baseUrl: "https://api.cerebras.ai/v1",
3022
- reasoning: false,
3005
+ reasoning: true,
3023
3006
  input: ["text"],
3024
3007
  cost: {
3025
3008
  input: 2.25,
@@ -3028,7 +3011,7 @@ export const MODELS = {
3028
3011
  cacheWrite: 0,
3029
3012
  },
3030
3013
  contextWindow: 131072,
3031
- maxTokens: 40000,
3014
+ maxTokens: 40960,
3032
3015
  },
3033
3016
  },
3034
3017
  "cloudflare-ai-gateway": {
@@ -4159,6 +4142,25 @@ export const MODELS = {
4159
4142
  },
4160
4143
  },
4161
4144
  "github-copilot": {
4145
+ "claude-fable-5": {
4146
+ id: "claude-fable-5",
4147
+ name: "Claude Fable 5",
4148
+ api: "openai-completions",
4149
+ provider: "github-copilot",
4150
+ baseUrl: "https://api.individual.githubcopilot.com",
4151
+ 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" },
4152
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4153
+ reasoning: true,
4154
+ input: ["text", "image"],
4155
+ cost: {
4156
+ input: 10,
4157
+ output: 50,
4158
+ cacheRead: 1,
4159
+ cacheWrite: 12.5,
4160
+ },
4161
+ contextWindow: 1000000,
4162
+ maxTokens: 128000,
4163
+ },
4162
4164
  "claude-haiku-4.5": {
4163
4165
  id: "claude-haiku-4.5",
4164
4166
  name: "Claude Haiku 4.5 (latest)",
@@ -6808,25 +6810,6 @@ export const MODELS = {
6808
6810
  contextWindow: 131072,
6809
6811
  maxTokens: 32768,
6810
6812
  },
6811
- "qwen/qwen3-coder-480b-a35b-instruct": {
6812
- id: "qwen/qwen3-coder-480b-a35b-instruct",
6813
- name: "Qwen3 Coder 480B A35B Instruct",
6814
- api: "openai-completions",
6815
- provider: "nvidia",
6816
- baseUrl: "https://integrate.api.nvidia.com/v1",
6817
- headers: { "NVCF-POLL-SECONDS": "3600" },
6818
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6819
- reasoning: false,
6820
- input: ["text"],
6821
- cost: {
6822
- input: 0,
6823
- output: 0,
6824
- cacheRead: 0,
6825
- cacheWrite: 0,
6826
- },
6827
- contextWindow: 262144,
6828
- maxTokens: 66536,
6829
- },
6830
6813
  "qwen/qwen3.5-122b-a10b": {
6831
6814
  id: "qwen/qwen3.5-122b-a10b",
6832
6815
  name: "Qwen3.5 122B-A10B",
@@ -7929,7 +7912,7 @@ export const MODELS = {
7929
7912
  cost: {
7930
7913
  input: 0.14,
7931
7914
  output: 0.28,
7932
- cacheRead: 0.03,
7915
+ cacheRead: 0.028,
7933
7916
  cacheWrite: 0,
7934
7917
  },
7935
7918
  contextWindow: 1000000,
@@ -7954,6 +7937,25 @@ export const MODELS = {
7954
7937
  contextWindow: 200000,
7955
7938
  maxTokens: 128000,
7956
7939
  },
7940
+ "deepseek-v4-pro": {
7941
+ id: "deepseek-v4-pro",
7942
+ name: "DeepSeek V4 Pro",
7943
+ api: "openai-completions",
7944
+ provider: "opencode",
7945
+ baseUrl: "https://opencode.ai/zen/v1",
7946
+ compat: { "maxTokensField": "max_tokens", "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7947
+ reasoning: true,
7948
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7949
+ input: ["text"],
7950
+ cost: {
7951
+ input: 1.74,
7952
+ output: 3.84,
7953
+ cacheRead: 0.145,
7954
+ cacheWrite: 0,
7955
+ },
7956
+ contextWindow: 1000000,
7957
+ maxTokens: 384000,
7958
+ },
7957
7959
  "gemini-3-flash": {
7958
7960
  id: "gemini-3-flash",
7959
7961
  name: "Gemini 3 Flash",
@@ -9347,7 +9349,7 @@ export const MODELS = {
9347
9349
  cacheRead: 0.135,
9348
9350
  cacheWrite: 0,
9349
9351
  },
9350
- contextWindow: 163840,
9352
+ contextWindow: 131072,
9351
9353
  maxTokens: 16384,
9352
9354
  },
9353
9355
  "deepseek/deepseek-chat-v3.1": {
@@ -9463,13 +9465,13 @@ export const MODELS = {
9463
9465
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
9464
9466
  input: ["text"],
9465
9467
  cost: {
9466
- input: 0.0983,
9467
- output: 0.1966,
9468
- cacheRead: 0.019700000000000002,
9468
+ input: 0.098,
9469
+ output: 0.196,
9470
+ cacheRead: 0.02,
9469
9471
  cacheWrite: 0,
9470
9472
  },
9471
9473
  contextWindow: 1048576,
9472
- maxTokens: 131072,
9474
+ maxTokens: 4096,
9473
9475
  },
9474
9476
  "deepseek/deepseek-v4-pro": {
9475
9477
  id: "deepseek/deepseek-v4-pro",
@@ -9789,12 +9791,12 @@ export const MODELS = {
9789
9791
  input: ["text", "image"],
9790
9792
  cost: {
9791
9793
  input: 0.12,
9792
- output: 0.36,
9794
+ output: 0.35,
9793
9795
  cacheRead: 0.09,
9794
9796
  cacheWrite: 0,
9795
9797
  },
9796
9798
  contextWindow: 262144,
9797
- maxTokens: 8192,
9799
+ maxTokens: 262144,
9798
9800
  },
9799
9801
  "google/gemma-4-31b-it:free": {
9800
9802
  id: "google/gemma-4-31b-it:free",
@@ -10095,9 +10097,9 @@ export const MODELS = {
10095
10097
  reasoning: true,
10096
10098
  input: ["text"],
10097
10099
  cost: {
10098
- input: 0.27,
10099
- output: 1.08,
10100
- cacheRead: 0.054,
10100
+ input: 0.25,
10101
+ output: 1,
10102
+ cacheRead: 0.049999999999999996,
10101
10103
  cacheWrite: 0,
10102
10104
  },
10103
10105
  contextWindow: 204800,
@@ -10487,31 +10489,13 @@ export const MODELS = {
10487
10489
  reasoning: true,
10488
10490
  input: ["text", "image"],
10489
10491
  cost: {
10490
- input: 0.6799999999999999,
10491
- output: 3.41,
10492
- cacheRead: 0.33999999999999997,
10492
+ input: 0.67,
10493
+ output: 3.39,
10494
+ cacheRead: 0.14,
10493
10495
  cacheWrite: 0,
10494
10496
  },
10495
10497
  contextWindow: 262144,
10496
- maxTokens: 262142,
10497
- },
10498
- "moonshotai/kimi-k2.6:free": {
10499
- id: "moonshotai/kimi-k2.6:free",
10500
- name: "MoonshotAI: Kimi K2.6 (free)",
10501
- api: "openai-completions",
10502
- provider: "openrouter",
10503
- baseUrl: "https://openrouter.ai/api/v1",
10504
- compat: { "supportsDeveloperRole": false, "requiresReasoningContentOnAssistantMessages": true },
10505
- reasoning: true,
10506
- input: ["text", "image"],
10507
- cost: {
10508
- input: 0,
10509
- output: 0,
10510
- cacheRead: 0,
10511
- cacheWrite: 0,
10512
- },
10513
- contextWindow: 262144,
10514
- maxTokens: 4096,
10498
+ maxTokens: 262144,
10515
10499
  },
10516
10500
  "nex-agi/nex-n2-pro:free": {
10517
10501
  id: "nex-agi/nex-n2-pro:free",
@@ -10683,23 +10667,6 @@ export const MODELS = {
10683
10667
  contextWindow: 128000,
10684
10668
  maxTokens: 128000,
10685
10669
  },
10686
- "nvidia/nemotron-nano-9b-v2": {
10687
- id: "nvidia/nemotron-nano-9b-v2",
10688
- name: "NVIDIA: Nemotron Nano 9B V2",
10689
- api: "openai-completions",
10690
- provider: "openrouter",
10691
- baseUrl: "https://openrouter.ai/api/v1",
10692
- reasoning: true,
10693
- input: ["text"],
10694
- cost: {
10695
- input: 0.04,
10696
- output: 0.16,
10697
- cacheRead: 0,
10698
- cacheWrite: 0,
10699
- },
10700
- contextWindow: 131072,
10701
- maxTokens: 16384,
10702
- },
10703
10670
  "nvidia/nemotron-nano-9b-v2:free": {
10704
10671
  id: "nvidia/nemotron-nano-9b-v2:free",
10705
10672
  name: "NVIDIA: Nemotron Nano 9B V2 (free)",
@@ -12438,13 +12405,13 @@ export const MODELS = {
12438
12405
  reasoning: true,
12439
12406
  input: ["text", "image"],
12440
12407
  cost: {
12441
- input: 0.14,
12408
+ input: 0.15,
12442
12409
  output: 1,
12443
- cacheRead: 0,
12410
+ cacheRead: 0.049999999999999996,
12444
12411
  cacheWrite: 0,
12445
12412
  },
12446
12413
  contextWindow: 262144,
12447
- maxTokens: 262140,
12414
+ maxTokens: 262144,
12448
12415
  },
12449
12416
  "qwen/qwen3.6-flash": {
12450
12417
  id: "qwen/qwen3.6-flash",
@@ -12523,10 +12490,10 @@ export const MODELS = {
12523
12490
  reasoning: true,
12524
12491
  input: ["text", "image"],
12525
12492
  cost: {
12526
- input: 0.39999999999999997,
12527
- output: 1.5999999999999999,
12528
- cacheRead: 0.08,
12529
- cacheWrite: 0.5,
12493
+ input: 0.32,
12494
+ output: 1.28,
12495
+ cacheRead: 0.064,
12496
+ cacheWrite: 0.39999999999999997,
12530
12497
  },
12531
12498
  contextWindow: 1000000,
12532
12499
  maxTokens: 65536,
@@ -13067,13 +13034,13 @@ export const MODELS = {
13067
13034
  reasoning: true,
13068
13035
  input: ["text", "image"],
13069
13036
  cost: {
13070
- input: 0.6799999999999999,
13071
- output: 3.41,
13072
- cacheRead: 0.33999999999999997,
13037
+ input: 0.67,
13038
+ output: 3.39,
13039
+ cacheRead: 0.14,
13073
13040
  cacheWrite: 0,
13074
13041
  },
13075
13042
  contextWindow: 262144,
13076
- maxTokens: 262142,
13043
+ maxTokens: 262144,
13077
13044
  },
13078
13045
  "~openai/gpt-latest": {
13079
13046
  id: "~openai/gpt-latest",
@@ -13111,25 +13078,6 @@ export const MODELS = {
13111
13078
  },
13112
13079
  },
13113
13080
  "together": {
13114
- "MiniMaxAI/MiniMax-M2.5": {
13115
- id: "MiniMaxAI/MiniMax-M2.5",
13116
- name: "MiniMax-M2.5",
13117
- api: "openai-completions",
13118
- provider: "together",
13119
- baseUrl: "https://api.together.ai/v1",
13120
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13121
- reasoning: true,
13122
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13123
- input: ["text"],
13124
- cost: {
13125
- input: 0.3,
13126
- output: 1.2,
13127
- cacheRead: 0.06,
13128
- cacheWrite: 0,
13129
- },
13130
- contextWindow: 204800,
13131
- maxTokens: 131072,
13132
- },
13133
13081
  "MiniMaxAI/MiniMax-M2.7": {
13134
13082
  id: "MiniMaxAI/MiniMax-M2.7",
13135
13083
  name: "MiniMax-M2.7",
@@ -13149,28 +13097,27 @@ export const MODELS = {
13149
13097
  contextWindow: 202752,
13150
13098
  maxTokens: 131072,
13151
13099
  },
13152
- "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
13153
- id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
13154
- name: "Qwen3 235B A22B Instruct 2507 FP8",
13100
+ "Qwen/Qwen2.5-7B-Instruct-Turbo": {
13101
+ id: "Qwen/Qwen2.5-7B-Instruct-Turbo",
13102
+ name: "Qwen 2.5 7B Instruct Turbo",
13155
13103
  api: "openai-completions",
13156
13104
  provider: "together",
13157
13105
  baseUrl: "https://api.together.ai/v1",
13158
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13159
- reasoning: true,
13160
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13106
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
13107
+ reasoning: false,
13161
13108
  input: ["text"],
13162
13109
  cost: {
13163
- input: 0.2,
13164
- output: 0.6,
13110
+ input: 0.3,
13111
+ output: 0.3,
13165
13112
  cacheRead: 0,
13166
13113
  cacheWrite: 0,
13167
13114
  },
13168
- contextWindow: 262144,
13169
- maxTokens: 262144,
13115
+ contextWindow: 32768,
13116
+ maxTokens: 32768,
13170
13117
  },
13171
- "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": {
13172
- id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8",
13173
- name: "Qwen3 Coder 480B A35B Instruct",
13118
+ "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
13119
+ id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
13120
+ name: "Qwen3 235B A22B Instruct 2507 FP8",
13174
13121
  api: "openai-completions",
13175
13122
  provider: "together",
13176
13123
  baseUrl: "https://api.together.ai/v1",
@@ -13178,36 +13125,36 @@ export const MODELS = {
13178
13125
  reasoning: false,
13179
13126
  input: ["text"],
13180
13127
  cost: {
13181
- input: 2,
13182
- output: 2,
13128
+ input: 0.2,
13129
+ output: 0.6,
13183
13130
  cacheRead: 0,
13184
13131
  cacheWrite: 0,
13185
13132
  },
13186
13133
  contextWindow: 262144,
13187
13134
  maxTokens: 262144,
13188
13135
  },
13189
- "Qwen/Qwen3-Coder-Next-FP8": {
13190
- id: "Qwen/Qwen3-Coder-Next-FP8",
13191
- name: "Qwen3 Coder Next FP8",
13136
+ "Qwen/Qwen3.5-397B-A17B": {
13137
+ id: "Qwen/Qwen3.5-397B-A17B",
13138
+ name: "Qwen3.5 397B A17B",
13192
13139
  api: "openai-completions",
13193
13140
  provider: "together",
13194
13141
  baseUrl: "https://api.together.ai/v1",
13195
13142
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13196
13143
  reasoning: true,
13197
13144
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13198
- input: ["text"],
13145
+ input: ["text", "image"],
13199
13146
  cost: {
13200
- input: 0.5,
13201
- output: 1.2,
13147
+ input: 0.6,
13148
+ output: 3.6,
13202
13149
  cacheRead: 0,
13203
13150
  cacheWrite: 0,
13204
13151
  },
13205
13152
  contextWindow: 262144,
13206
- maxTokens: 262144,
13153
+ maxTokens: 130000,
13207
13154
  },
13208
- "Qwen/Qwen3.5-397B-A17B": {
13209
- id: "Qwen/Qwen3.5-397B-A17B",
13210
- name: "Qwen3.5 397B A17B",
13155
+ "Qwen/Qwen3.5-9B": {
13156
+ id: "Qwen/Qwen3.5-9B",
13157
+ name: "Qwen3.5 9B",
13211
13158
  api: "openai-completions",
13212
13159
  provider: "together",
13213
13160
  baseUrl: "https://api.together.ai/v1",
@@ -13216,13 +13163,13 @@ export const MODELS = {
13216
13163
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13217
13164
  input: ["text", "image"],
13218
13165
  cost: {
13219
- input: 0.6,
13220
- output: 3.6,
13166
+ input: 0.17,
13167
+ output: 0.25,
13221
13168
  cacheRead: 0,
13222
13169
  cacheWrite: 0,
13223
13170
  },
13224
13171
  contextWindow: 262144,
13225
- maxTokens: 130000,
13172
+ maxTokens: 65536,
13226
13173
  },
13227
13174
  "Qwen/Qwen3.6-Plus": {
13228
13175
  id: "Qwen/Qwen3.6-Plus",
@@ -13249,9 +13196,8 @@ export const MODELS = {
13249
13196
  api: "openai-completions",
13250
13197
  provider: "together",
13251
13198
  baseUrl: "https://api.together.ai/v1",
13252
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13253
- reasoning: true,
13254
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13199
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
13200
+ reasoning: false,
13255
13201
  input: ["text"],
13256
13202
  cost: {
13257
13203
  input: 2.5,
@@ -13262,44 +13208,6 @@ export const MODELS = {
13262
13208
  contextWindow: 1000000,
13263
13209
  maxTokens: 500000,
13264
13210
  },
13265
- "deepseek-ai/DeepSeek-V3": {
13266
- id: "deepseek-ai/DeepSeek-V3",
13267
- name: "DeepSeek-V3",
13268
- api: "openai-completions",
13269
- provider: "together",
13270
- baseUrl: "https://api.together.ai/v1",
13271
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13272
- reasoning: true,
13273
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13274
- input: ["text"],
13275
- cost: {
13276
- input: 1.25,
13277
- output: 1.25,
13278
- cacheRead: 0,
13279
- cacheWrite: 0,
13280
- },
13281
- contextWindow: 131072,
13282
- maxTokens: 131072,
13283
- },
13284
- "deepseek-ai/DeepSeek-V3-1": {
13285
- id: "deepseek-ai/DeepSeek-V3-1",
13286
- name: "DeepSeek V3.1",
13287
- api: "openai-completions",
13288
- provider: "together",
13289
- baseUrl: "https://api.together.ai/v1",
13290
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13291
- reasoning: true,
13292
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13293
- input: ["text"],
13294
- cost: {
13295
- input: 0.6,
13296
- output: 1.7,
13297
- cacheRead: 0,
13298
- cacheWrite: 0,
13299
- },
13300
- contextWindow: 131072,
13301
- maxTokens: 131072,
13302
- },
13303
13211
  "deepseek-ai/DeepSeek-V4-Pro": {
13304
13212
  id: "deepseek-ai/DeepSeek-V4-Pro",
13305
13213
  name: "DeepSeek V4 Pro",
@@ -13311,8 +13219,8 @@ export const MODELS = {
13311
13219
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": null },
13312
13220
  input: ["text"],
13313
13221
  cost: {
13314
- input: 2.1,
13315
- output: 4.4,
13222
+ input: 1.74,
13223
+ output: 3.48,
13316
13224
  cacheRead: 0.2,
13317
13225
  cacheWrite: 0,
13318
13226
  },
@@ -13348,8 +13256,8 @@ export const MODELS = {
13348
13256
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13349
13257
  input: ["text", "image"],
13350
13258
  cost: {
13351
- input: 0.2,
13352
- output: 0.5,
13259
+ input: 0.39,
13260
+ output: 0.97,
13353
13261
  cacheRead: 0,
13354
13262
  cacheWrite: 0,
13355
13263
  },
@@ -13374,25 +13282,6 @@ export const MODELS = {
13374
13282
  contextWindow: 131072,
13375
13283
  maxTokens: 131072,
13376
13284
  },
13377
- "moonshotai/Kimi-K2.5": {
13378
- id: "moonshotai/Kimi-K2.5",
13379
- name: "Kimi K2.5",
13380
- api: "openai-completions",
13381
- provider: "together",
13382
- baseUrl: "https://api.together.ai/v1",
13383
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13384
- reasoning: true,
13385
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13386
- input: ["text", "image"],
13387
- cost: {
13388
- input: 0.5,
13389
- output: 2.8,
13390
- cacheRead: 0,
13391
- cacheWrite: 0,
13392
- },
13393
- contextWindow: 262144,
13394
- maxTokens: 262144,
13395
- },
13396
13285
  "moonshotai/Kimi-K2.6": {
13397
13286
  id: "moonshotai/Kimi-K2.6",
13398
13287
  name: "Kimi K2.6",
@@ -13450,6 +13339,44 @@ export const MODELS = {
13450
13339
  contextWindow: 131072,
13451
13340
  maxTokens: 131072,
13452
13341
  },
13342
+ "openai/gpt-oss-20b": {
13343
+ id: "openai/gpt-oss-20b",
13344
+ name: "GPT OSS 20B",
13345
+ api: "openai-completions",
13346
+ provider: "together",
13347
+ baseUrl: "https://api.together.ai/v1",
13348
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": true, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "openai" },
13349
+ reasoning: true,
13350
+ thinkingLevelMap: { "off": null, "minimal": null },
13351
+ input: ["text"],
13352
+ cost: {
13353
+ input: 0.05,
13354
+ output: 0.2,
13355
+ cacheRead: 0,
13356
+ cacheWrite: 0,
13357
+ },
13358
+ contextWindow: 131072,
13359
+ maxTokens: 131072,
13360
+ },
13361
+ "zai-org/GLM-5": {
13362
+ id: "zai-org/GLM-5",
13363
+ name: "GLM-5",
13364
+ api: "openai-completions",
13365
+ provider: "together",
13366
+ baseUrl: "https://api.together.ai/v1",
13367
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13368
+ reasoning: true,
13369
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13370
+ input: ["text"],
13371
+ cost: {
13372
+ input: 1,
13373
+ output: 3.2,
13374
+ cacheRead: 0,
13375
+ cacheWrite: 0,
13376
+ },
13377
+ contextWindow: 202752,
13378
+ maxTokens: 131072,
13379
+ },
13453
13380
  "zai-org/GLM-5.1": {
13454
13381
  id: "zai-org/GLM-5.1",
13455
13382
  name: "GLM-5.1",
@@ -15027,40 +14954,6 @@ export const MODELS = {
15027
14954
  contextWindow: 262114,
15028
14955
  maxTokens: 262114,
15029
14956
  },
15030
- "moonshotai/kimi-k2-thinking-turbo": {
15031
- id: "moonshotai/kimi-k2-thinking-turbo",
15032
- name: "Kimi K2 Thinking Turbo",
15033
- api: "anthropic-messages",
15034
- provider: "vercel-ai-gateway",
15035
- baseUrl: "https://ai-gateway.vercel.sh",
15036
- reasoning: true,
15037
- input: ["text"],
15038
- cost: {
15039
- input: 1.15,
15040
- output: 8,
15041
- cacheRead: 0.15,
15042
- cacheWrite: 0,
15043
- },
15044
- contextWindow: 262114,
15045
- maxTokens: 262114,
15046
- },
15047
- "moonshotai/kimi-k2-turbo": {
15048
- id: "moonshotai/kimi-k2-turbo",
15049
- name: "Kimi K2 Turbo",
15050
- api: "anthropic-messages",
15051
- provider: "vercel-ai-gateway",
15052
- baseUrl: "https://ai-gateway.vercel.sh",
15053
- reasoning: false,
15054
- input: ["text"],
15055
- cost: {
15056
- input: 1.15,
15057
- output: 8,
15058
- cacheRead: 0.15,
15059
- cacheWrite: 0,
15060
- },
15061
- contextWindow: 256000,
15062
- maxTokens: 16384,
15063
- },
15064
14957
  "moonshotai/kimi-k2.5": {
15065
14958
  id: "moonshotai/kimi-k2.5",
15066
14959
  name: "Kimi K2.5",