@fleetagent/pi-ai 0.0.10 → 0.0.11

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.
@@ -2847,30 +2847,13 @@ export const MODELS = {
2847
2847
  reasoning: true,
2848
2848
  input: ["text"],
2849
2849
  cost: {
2850
- input: 0.25,
2851
- output: 0.69,
2850
+ input: 0.35,
2851
+ output: 0.75,
2852
2852
  cacheRead: 0,
2853
2853
  cacheWrite: 0,
2854
2854
  },
2855
2855
  contextWindow: 131072,
2856
- maxTokens: 32768,
2857
- },
2858
- "llama3.1-8b": {
2859
- id: "llama3.1-8b",
2860
- name: "Llama 3.1 8B",
2861
- api: "openai-completions",
2862
- provider: "cerebras",
2863
- baseUrl: "https://api.cerebras.ai/v1",
2864
- reasoning: false,
2865
- input: ["text"],
2866
- cost: {
2867
- input: 0.1,
2868
- output: 0.1,
2869
- cacheRead: 0,
2870
- cacheWrite: 0,
2871
- },
2872
- contextWindow: 32000,
2873
- maxTokens: 8000,
2856
+ maxTokens: 40960,
2874
2857
  },
2875
2858
  "zai-glm-4.7": {
2876
2859
  id: "zai-glm-4.7",
@@ -2878,7 +2861,7 @@ export const MODELS = {
2878
2861
  api: "openai-completions",
2879
2862
  provider: "cerebras",
2880
2863
  baseUrl: "https://api.cerebras.ai/v1",
2881
- reasoning: false,
2864
+ reasoning: true,
2882
2865
  input: ["text"],
2883
2866
  cost: {
2884
2867
  input: 2.25,
@@ -2887,7 +2870,7 @@ export const MODELS = {
2887
2870
  cacheWrite: 0,
2888
2871
  },
2889
2872
  contextWindow: 131072,
2890
- maxTokens: 40000,
2873
+ maxTokens: 40960,
2891
2874
  },
2892
2875
  },
2893
2876
  "cloudflare-ai-gateway": {
@@ -4015,6 +3998,25 @@ export const MODELS = {
4015
3998
  },
4016
3999
  },
4017
4000
  "github-copilot": {
4001
+ "claude-fable-5": {
4002
+ id: "claude-fable-5",
4003
+ name: "Claude Fable 5",
4004
+ api: "openai-completions",
4005
+ provider: "github-copilot",
4006
+ baseUrl: "https://api.individual.githubcopilot.com",
4007
+ 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" },
4008
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4009
+ reasoning: true,
4010
+ input: ["text", "image"],
4011
+ cost: {
4012
+ input: 10,
4013
+ output: 50,
4014
+ cacheRead: 1,
4015
+ cacheWrite: 12.5,
4016
+ },
4017
+ contextWindow: 1000000,
4018
+ maxTokens: 128000,
4019
+ },
4018
4020
  "claude-haiku-4.5": {
4019
4021
  id: "claude-haiku-4.5",
4020
4022
  name: "Claude Haiku 4.5 (latest)",
@@ -9263,12 +9265,12 @@ export const MODELS = {
9263
9265
  input: ["text", "image"],
9264
9266
  cost: {
9265
9267
  input: 0.12,
9266
- output: 0.36,
9268
+ output: 0.35,
9267
9269
  cacheRead: 0.09,
9268
9270
  cacheWrite: 0,
9269
9271
  },
9270
9272
  contextWindow: 262144,
9271
- maxTokens: 8192,
9273
+ maxTokens: 262144,
9272
9274
  },
9273
9275
  "google/gemma-4-31b-it:free": {
9274
9276
  id: "google/gemma-4-31b-it:free",
@@ -9569,9 +9571,9 @@ export const MODELS = {
9569
9571
  reasoning: true,
9570
9572
  input: ["text"],
9571
9573
  cost: {
9572
- input: 0.27,
9573
- output: 1.08,
9574
- cacheRead: 0.054,
9574
+ input: 0.25,
9575
+ output: 1,
9576
+ cacheRead: 0.049999999999999996,
9575
9577
  cacheWrite: 0,
9576
9578
  },
9577
9579
  contextWindow: 204800,
@@ -9961,13 +9963,13 @@ export const MODELS = {
9961
9963
  reasoning: true,
9962
9964
  input: ["text", "image"],
9963
9965
  cost: {
9964
- input: 0.6799999999999999,
9965
- output: 3.41,
9966
- cacheRead: 0.33999999999999997,
9966
+ input: 0.67,
9967
+ output: 3.39,
9968
+ cacheRead: 0.14,
9967
9969
  cacheWrite: 0,
9968
9970
  },
9969
9971
  contextWindow: 262144,
9970
- maxTokens: 262142,
9972
+ maxTokens: 262144,
9971
9973
  },
9972
9974
  "nex-agi/nex-n2-pro:free": {
9973
9975
  id: "nex-agi/nex-n2-pro:free",
@@ -10139,23 +10141,6 @@ export const MODELS = {
10139
10141
  contextWindow: 128000,
10140
10142
  maxTokens: 128000,
10141
10143
  },
10142
- "nvidia/nemotron-nano-9b-v2": {
10143
- id: "nvidia/nemotron-nano-9b-v2",
10144
- name: "NVIDIA: Nemotron Nano 9B V2",
10145
- api: "openai-completions",
10146
- provider: "openrouter",
10147
- baseUrl: "https://openrouter.ai/api/v1",
10148
- reasoning: true,
10149
- input: ["text"],
10150
- cost: {
10151
- input: 0.04,
10152
- output: 0.16,
10153
- cacheRead: 0,
10154
- cacheWrite: 0,
10155
- },
10156
- contextWindow: 131072,
10157
- maxTokens: 16384,
10158
- },
10159
10144
  "nvidia/nemotron-nano-9b-v2:free": {
10160
10145
  id: "nvidia/nemotron-nano-9b-v2:free",
10161
10146
  name: "NVIDIA: Nemotron Nano 9B V2 (free)",
@@ -11979,10 +11964,10 @@ export const MODELS = {
11979
11964
  reasoning: true,
11980
11965
  input: ["text", "image"],
11981
11966
  cost: {
11982
- input: 0.39999999999999997,
11983
- output: 1.5999999999999999,
11984
- cacheRead: 0.08,
11985
- cacheWrite: 0.5,
11967
+ input: 0.32,
11968
+ output: 1.28,
11969
+ cacheRead: 0.064,
11970
+ cacheWrite: 0.39999999999999997,
11986
11971
  },
11987
11972
  contextWindow: 1000000,
11988
11973
  maxTokens: 65536,
@@ -12523,13 +12508,13 @@ export const MODELS = {
12523
12508
  reasoning: true,
12524
12509
  input: ["text", "image"],
12525
12510
  cost: {
12526
- input: 0.6799999999999999,
12527
- output: 3.41,
12528
- cacheRead: 0.33999999999999997,
12511
+ input: 0.67,
12512
+ output: 3.39,
12513
+ cacheRead: 0.14,
12529
12514
  cacheWrite: 0,
12530
12515
  },
12531
12516
  contextWindow: 262144,
12532
- maxTokens: 262142,
12517
+ maxTokens: 262144,
12533
12518
  },
12534
12519
  "~openai/gpt-latest": {
12535
12520
  id: "~openai/gpt-latest",
@@ -12567,25 +12552,6 @@ export const MODELS = {
12567
12552
  },
12568
12553
  },
12569
12554
  "together": {
12570
- "MiniMaxAI/MiniMax-M2.5": {
12571
- id: "MiniMaxAI/MiniMax-M2.5",
12572
- name: "MiniMax-M2.5",
12573
- api: "openai-completions",
12574
- provider: "together",
12575
- baseUrl: "https://api.together.ai/v1",
12576
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12577
- reasoning: true,
12578
- thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null },
12579
- input: ["text"],
12580
- cost: {
12581
- input: 0.3,
12582
- output: 1.2,
12583
- cacheRead: 0.06,
12584
- cacheWrite: 0,
12585
- },
12586
- contextWindow: 204800,
12587
- maxTokens: 131072,
12588
- },
12589
12555
  "MiniMaxAI/MiniMax-M2.7": {
12590
12556
  id: "MiniMaxAI/MiniMax-M2.7",
12591
12557
  name: "MiniMax-M2.7",
@@ -12605,28 +12571,27 @@ export const MODELS = {
12605
12571
  contextWindow: 202752,
12606
12572
  maxTokens: 131072,
12607
12573
  },
12608
- "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
12609
- id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
12610
- name: "Qwen3 235B A22B Instruct 2507 FP8",
12574
+ "Qwen/Qwen2.5-7B-Instruct-Turbo": {
12575
+ id: "Qwen/Qwen2.5-7B-Instruct-Turbo",
12576
+ name: "Qwen 2.5 7B Instruct Turbo",
12611
12577
  api: "openai-completions",
12612
12578
  provider: "together",
12613
12579
  baseUrl: "https://api.together.ai/v1",
12614
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12615
- reasoning: true,
12616
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12580
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12581
+ reasoning: false,
12617
12582
  input: ["text"],
12618
12583
  cost: {
12619
- input: 0.2,
12620
- output: 0.6,
12584
+ input: 0.3,
12585
+ output: 0.3,
12621
12586
  cacheRead: 0,
12622
12587
  cacheWrite: 0,
12623
12588
  },
12624
- contextWindow: 262144,
12625
- maxTokens: 262144,
12589
+ contextWindow: 32768,
12590
+ maxTokens: 32768,
12626
12591
  },
12627
- "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": {
12628
- id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8",
12629
- name: "Qwen3 Coder 480B A35B Instruct",
12592
+ "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
12593
+ id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
12594
+ name: "Qwen3 235B A22B Instruct 2507 FP8",
12630
12595
  api: "openai-completions",
12631
12596
  provider: "together",
12632
12597
  baseUrl: "https://api.together.ai/v1",
@@ -12634,36 +12599,36 @@ export const MODELS = {
12634
12599
  reasoning: false,
12635
12600
  input: ["text"],
12636
12601
  cost: {
12637
- input: 2,
12638
- output: 2,
12602
+ input: 0.2,
12603
+ output: 0.6,
12639
12604
  cacheRead: 0,
12640
12605
  cacheWrite: 0,
12641
12606
  },
12642
12607
  contextWindow: 262144,
12643
12608
  maxTokens: 262144,
12644
12609
  },
12645
- "Qwen/Qwen3-Coder-Next-FP8": {
12646
- id: "Qwen/Qwen3-Coder-Next-FP8",
12647
- name: "Qwen3 Coder Next FP8",
12610
+ "Qwen/Qwen3.5-397B-A17B": {
12611
+ id: "Qwen/Qwen3.5-397B-A17B",
12612
+ name: "Qwen3.5 397B A17B",
12648
12613
  api: "openai-completions",
12649
12614
  provider: "together",
12650
12615
  baseUrl: "https://api.together.ai/v1",
12651
12616
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12652
12617
  reasoning: true,
12653
12618
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12654
- input: ["text"],
12619
+ input: ["text", "image"],
12655
12620
  cost: {
12656
- input: 0.5,
12657
- output: 1.2,
12621
+ input: 0.6,
12622
+ output: 3.6,
12658
12623
  cacheRead: 0,
12659
12624
  cacheWrite: 0,
12660
12625
  },
12661
12626
  contextWindow: 262144,
12662
- maxTokens: 262144,
12627
+ maxTokens: 130000,
12663
12628
  },
12664
- "Qwen/Qwen3.5-397B-A17B": {
12665
- id: "Qwen/Qwen3.5-397B-A17B",
12666
- name: "Qwen3.5 397B A17B",
12629
+ "Qwen/Qwen3.5-9B": {
12630
+ id: "Qwen/Qwen3.5-9B",
12631
+ name: "Qwen3.5 9B",
12667
12632
  api: "openai-completions",
12668
12633
  provider: "together",
12669
12634
  baseUrl: "https://api.together.ai/v1",
@@ -12672,13 +12637,13 @@ export const MODELS = {
12672
12637
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12673
12638
  input: ["text", "image"],
12674
12639
  cost: {
12675
- input: 0.6,
12676
- output: 3.6,
12640
+ input: 0.17,
12641
+ output: 0.25,
12677
12642
  cacheRead: 0,
12678
12643
  cacheWrite: 0,
12679
12644
  },
12680
12645
  contextWindow: 262144,
12681
- maxTokens: 130000,
12646
+ maxTokens: 65536,
12682
12647
  },
12683
12648
  "Qwen/Qwen3.6-Plus": {
12684
12649
  id: "Qwen/Qwen3.6-Plus",
@@ -12705,9 +12670,8 @@ export const MODELS = {
12705
12670
  api: "openai-completions",
12706
12671
  provider: "together",
12707
12672
  baseUrl: "https://api.together.ai/v1",
12708
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12709
- reasoning: true,
12710
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12673
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12674
+ reasoning: false,
12711
12675
  input: ["text"],
12712
12676
  cost: {
12713
12677
  input: 2.5,
@@ -12718,44 +12682,6 @@ export const MODELS = {
12718
12682
  contextWindow: 1000000,
12719
12683
  maxTokens: 500000,
12720
12684
  },
12721
- "deepseek-ai/DeepSeek-V3": {
12722
- id: "deepseek-ai/DeepSeek-V3",
12723
- name: "DeepSeek-V3",
12724
- api: "openai-completions",
12725
- provider: "together",
12726
- baseUrl: "https://api.together.ai/v1",
12727
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12728
- reasoning: true,
12729
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12730
- input: ["text"],
12731
- cost: {
12732
- input: 1.25,
12733
- output: 1.25,
12734
- cacheRead: 0,
12735
- cacheWrite: 0,
12736
- },
12737
- contextWindow: 131072,
12738
- maxTokens: 131072,
12739
- },
12740
- "deepseek-ai/DeepSeek-V3-1": {
12741
- id: "deepseek-ai/DeepSeek-V3-1",
12742
- name: "DeepSeek V3.1",
12743
- api: "openai-completions",
12744
- provider: "together",
12745
- baseUrl: "https://api.together.ai/v1",
12746
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12747
- reasoning: true,
12748
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12749
- input: ["text"],
12750
- cost: {
12751
- input: 0.6,
12752
- output: 1.7,
12753
- cacheRead: 0,
12754
- cacheWrite: 0,
12755
- },
12756
- contextWindow: 131072,
12757
- maxTokens: 131072,
12758
- },
12759
12685
  "deepseek-ai/DeepSeek-V4-Pro": {
12760
12686
  id: "deepseek-ai/DeepSeek-V4-Pro",
12761
12687
  name: "DeepSeek V4 Pro",
@@ -12767,8 +12693,8 @@ export const MODELS = {
12767
12693
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": null },
12768
12694
  input: ["text"],
12769
12695
  cost: {
12770
- input: 2.1,
12771
- output: 4.4,
12696
+ input: 1.74,
12697
+ output: 3.48,
12772
12698
  cacheRead: 0.2,
12773
12699
  cacheWrite: 0,
12774
12700
  },
@@ -12804,8 +12730,8 @@ export const MODELS = {
12804
12730
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12805
12731
  input: ["text", "image"],
12806
12732
  cost: {
12807
- input: 0.2,
12808
- output: 0.5,
12733
+ input: 0.39,
12734
+ output: 0.97,
12809
12735
  cacheRead: 0,
12810
12736
  cacheWrite: 0,
12811
12737
  },
@@ -12830,25 +12756,6 @@ export const MODELS = {
12830
12756
  contextWindow: 131072,
12831
12757
  maxTokens: 131072,
12832
12758
  },
12833
- "moonshotai/Kimi-K2.5": {
12834
- id: "moonshotai/Kimi-K2.5",
12835
- name: "Kimi K2.5",
12836
- api: "openai-completions",
12837
- provider: "together",
12838
- baseUrl: "https://api.together.ai/v1",
12839
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12840
- reasoning: true,
12841
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12842
- input: ["text", "image"],
12843
- cost: {
12844
- input: 0.5,
12845
- output: 2.8,
12846
- cacheRead: 0,
12847
- cacheWrite: 0,
12848
- },
12849
- contextWindow: 262144,
12850
- maxTokens: 262144,
12851
- },
12852
12759
  "moonshotai/Kimi-K2.6": {
12853
12760
  id: "moonshotai/Kimi-K2.6",
12854
12761
  name: "Kimi K2.6",
@@ -12906,6 +12813,44 @@ export const MODELS = {
12906
12813
  contextWindow: 131072,
12907
12814
  maxTokens: 131072,
12908
12815
  },
12816
+ "openai/gpt-oss-20b": {
12817
+ id: "openai/gpt-oss-20b",
12818
+ name: "GPT OSS 20B",
12819
+ api: "openai-completions",
12820
+ provider: "together",
12821
+ baseUrl: "https://api.together.ai/v1",
12822
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": true, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "openai" },
12823
+ reasoning: true,
12824
+ thinkingLevelMap: { "off": null, "minimal": null },
12825
+ input: ["text"],
12826
+ cost: {
12827
+ input: 0.05,
12828
+ output: 0.2,
12829
+ cacheRead: 0,
12830
+ cacheWrite: 0,
12831
+ },
12832
+ contextWindow: 131072,
12833
+ maxTokens: 131072,
12834
+ },
12835
+ "zai-org/GLM-5": {
12836
+ id: "zai-org/GLM-5",
12837
+ name: "GLM-5",
12838
+ api: "openai-completions",
12839
+ provider: "together",
12840
+ baseUrl: "https://api.together.ai/v1",
12841
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12842
+ reasoning: true,
12843
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12844
+ input: ["text"],
12845
+ cost: {
12846
+ input: 1,
12847
+ output: 3.2,
12848
+ cacheRead: 0,
12849
+ cacheWrite: 0,
12850
+ },
12851
+ contextWindow: 202752,
12852
+ maxTokens: 131072,
12853
+ },
12909
12854
  "zai-org/GLM-5.1": {
12910
12855
  id: "zai-org/GLM-5.1",
12911
12856
  name: "GLM-5.1",
@@ -14480,40 +14425,6 @@ export const MODELS = {
14480
14425
  contextWindow: 262114,
14481
14426
  maxTokens: 262114,
14482
14427
  },
14483
- "moonshotai/kimi-k2-thinking-turbo": {
14484
- id: "moonshotai/kimi-k2-thinking-turbo",
14485
- name: "Kimi K2 Thinking Turbo",
14486
- api: "anthropic-messages",
14487
- provider: "vercel-ai-gateway",
14488
- baseUrl: "https://ai-gateway.vercel.sh",
14489
- reasoning: true,
14490
- input: ["text"],
14491
- cost: {
14492
- input: 1.15,
14493
- output: 8,
14494
- cacheRead: 0.15,
14495
- cacheWrite: 0,
14496
- },
14497
- contextWindow: 262114,
14498
- maxTokens: 262114,
14499
- },
14500
- "moonshotai/kimi-k2-turbo": {
14501
- id: "moonshotai/kimi-k2-turbo",
14502
- name: "Kimi K2 Turbo",
14503
- api: "anthropic-messages",
14504
- provider: "vercel-ai-gateway",
14505
- baseUrl: "https://ai-gateway.vercel.sh",
14506
- reasoning: false,
14507
- input: ["text"],
14508
- cost: {
14509
- input: 1.15,
14510
- output: 8,
14511
- cacheRead: 0.15,
14512
- cacheWrite: 0,
14513
- },
14514
- contextWindow: 256000,
14515
- maxTokens: 16384,
14516
- },
14517
14428
  "moonshotai/kimi-k2.5": {
14518
14429
  id: "moonshotai/kimi-k2.5",
14519
14430
  name: "Kimi K2.5",