vtcode_config/models/
model_id.rs

1use serde::{Deserialize, Serialize};
2
3mod as_str;
4mod capabilities;
5mod collection;
6mod defaults;
7mod description;
8mod display;
9mod format;
10mod openrouter;
11mod parse;
12mod provider;
13
14pub use capabilities::{
15    ModelCatalogEntry, ModelPricing, catalog_provider_keys, model_catalog_entry,
16    supported_models_for_provider,
17};
18
19/// Centralized enum for all supported model identifiers
20#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
21#[derive(Clone, Copy, Debug, Default, PartialEq, Eq, Hash, Serialize, Deserialize)]
22pub enum ModelId {
23    // Gemini models
24    /// Gemini 3.1 Pro Preview - Latest Gemini 3.1 Pro flagship
25    Gemini31ProPreview,
26    /// Gemini 3.1 Pro Preview Custom Tools - Optimized for custom tools & bash
27    Gemini31ProPreviewCustomTools,
28    /// Gemini 3.1 Flash Lite Preview - Most cost-efficient model, offering fastest performance for high-frequency, lightweight tasks
29    Gemini31FlashLitePreview,
30    /// Gemini 3.5 Flash - High-efficiency frontier model for fast inference
31    Gemini35Flash,
32    /// Gemini 3 Flash Preview - Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding
33    #[default]
34    Gemini3FlashPreview,
35
36    // OpenAI models
37    /// GPT-5.5 - Next-gen OpenAI model dated release (2026-04-23)
38    GPT55,
39    /// GPT-5 - Latest most capable OpenAI model (2025-08-07)
40    GPT5,
41    /// GPT-5.2 - Latest flagship general-purpose OpenAI model (2025-12-11)
42    GPT52,
43    /// GPT-5.2 Codex - Code-focused GPT-5.2 variant optimized for agentic coding with xhigh reasoning support
44    GPT52Codex,
45    /// GPT-5.4 - Mainline frontier GPT model for general-purpose and coding work
46    GPT54,
47    /// GPT-5.4 Pro - Higher-compute GPT-5.4 variant for difficult problems
48    GPT54Pro,
49    /// GPT-5.4 Nano - Lightweight GPT-5.4 variant optimized for speed and cost-efficiency
50    GPT54Nano,
51    /// GPT-5.4 Mini - Compact GPT-5.4 variant for cost-effective tasks with reduced reasoning overhead
52    GPT54Mini,
53    /// GPT-5.3 Codex - Code-focused GPT-5.3 variant optimized for agentic coding with reasoning effort support (low, medium, high, xhigh)
54    GPT53Codex,
55    /// GPT-5.1 Codex - Code-focused GPT-5.1 variant optimized for agentic coding
56    GPT51Codex,
57    /// GPT-5.1 Codex Max - Higher-compute GPT-5.1 Codex variant for longer-running engineering tasks
58    GPT51CodexMax,
59    /// GPT-5 Mini - Latest efficient OpenAI model (2025-08-07)
60    GPT5Mini,
61    /// GPT-5 Nano - Latest most cost-effective OpenAI model (2025-08-07)
62    GPT5Nano,
63    /// GPT-5 Codex - Code-focused GPT-5 variant optimized for agentic coding
64    GPT5Codex,
65    /// GPT-OSS 20B - OpenAI's open-source 20B parameter model using harmony
66    OpenAIGptOss20b,
67    /// GPT-OSS 120B - OpenAI's open-source 120B parameter model using harmony
68    OpenAIGptOss120b,
69
70    // Anthropic models
71    /// Claude Opus 4.8 - Anthropic's most capable model for complex reasoning and agentic coding
72    ClaudeOpus48,
73    /// Claude Opus 4.7 - Previous flagship Anthropic model with adaptive thinking
74    ClaudeOpus47,
75    /// Claude Opus 4.6 - Previous Anthropic flagship retained for compatibility
76    ClaudeOpus46,
77    /// Claude Sonnet 4.6 - Balanced flagship Anthropic model in VT Code's conservative rollout
78    ClaudeSonnet46,
79    /// Claude Haiku 4.5 - Latest efficient Anthropic model (2025-10-15)
80    ClaudeHaiku45,
81    /// Claude Mythos Preview - Invitation-only Anthropic research preview for defensive cybersecurity workflows
82    ClaudeMythosPreview,
83    /// GitHub Copilot auto model selection
84    CopilotAuto,
85    /// GitHub Copilot GPT-5.2 Codex
86    CopilotGPT52Codex,
87    /// GitHub Copilot GPT-5.1 Codex Max
88    CopilotGPT51CodexMax,
89    /// GitHub Copilot GPT-5.4
90    CopilotGPT54,
91    /// GitHub Copilot GPT-5.4 Mini
92    CopilotGPT54Mini,
93    /// GitHub Copilot Claude Sonnet 4.6
94    CopilotClaudeSonnet46,
95
96    // DeepSeek models
97    /// DeepSeek V4 Pro - High-performance reasoning model with advanced thinking
98    DeepSeekV4Pro,
99    /// DeepSeek V4 Flash - Fast inference model for cost-effective reasoning
100    DeepSeekV4Flash,
101
102    // Mistral AI models
103    /// Mistral Large 3 - State-of-the-art open-weight general-purpose multimodal model
104    MistralLarge3,
105    // Hugging Face models
106    /// DeepSeek V3.2 via Hugging Face router
107    HuggingFaceDeepseekV32,
108    /// OpenAI GPT-OSS 20B via Hugging Face router
109    HuggingFaceOpenAIGptOss20b,
110    /// OpenAI GPT-OSS 120B via Hugging Face router
111    HuggingFaceOpenAIGptOss120b,
112    /// DeepSeek V3.2 via Novita on Hugging Face router
113    HuggingFaceDeepseekV32Novita,
114    /// Xiaomi MiMo-V2-Flash via Novita on Hugging Face router
115    HuggingFaceXiaomiMimoV2FlashNovita,
116    /// MiniMax M2.5 via Novita on Hugging Face router
117    HuggingFaceMinimaxM25Novita,
118    /// Z.AI GLM-5 via Novita on Hugging Face router
119    HuggingFaceGlm5Novita,
120    /// Z.AI GLM-5.1 via zai-org provider on Hugging Face router
121    HuggingFaceGlm51ZaiOrg,
122    /// Qwen3-Coder-Next via Novita inference provider on Hugging Face router
123    HuggingFaceQwen3CoderNextNovita,
124    /// Qwen3.5-397B-A17B via Together inference provider on Hugging Face router
125    HuggingFaceQwen35397BA17BTogether,
126    /// Kimi K2.6 via Novita on Hugging Face router
127    HuggingFaceKimiK26Novita,
128    /// DeepSeek V4 Flash via Novita on Hugging Face router
129    HuggingFaceDeepseekV4FlashNovita,
130    /// DeepSeek V4 Pro via Together on Hugging Face router
131    HuggingFaceDeepseekV4ProTogether,
132    /// Step 3.5 Flash via Hugging Face router
133    HuggingFaceStep35Flash,
134
135    // StepFun models
136    /// Step 3.7 Flash - StepFun's flagship multimodal reasoning model with tool calling
137    StepFun37Flash,
138
139    /// GLM-5 - Flagship Z.ai foundation model for complex systems
140    ZaiGlm5,
141    /// GLM-5.1 - Next-gen Z.ai foundation model with improved reasoning
142    ZaiGlm51,
143
144    // MiMo models
145    /// MiMo V2.5 Pro - Xiaomi's flagship reasoning model with 1M context
146    MiMoV25Pro,
147    /// MiMo V2.5 - Xiaomi's omni-modal model with full-modal understanding and 1M context
148    MiMoV25,
149    /// MiMo V2 Flash - Xiaomi's lightweight fast model for high-throughput tasks
150    MiMoV2Flash,
151
152    // Moonshot models
153    /// Kimi K2.6 - Moonshot.ai's latest 1T MoE flagship (32B active, MLA, MoonViT vision)
154    MoonshotKimiK26,
155    /// Kimi K2.5 - Moonshot.ai's previous flagship reasoning model
156    MoonshotKimiK25,
157
158    // OpenCode Zen models
159    /// GPT-5.4 - OpenCode Zen default flagship model
160    OpenCodeZenGPT54,
161    /// GPT-5.4 Mini - Lower-cost OpenCode Zen GPT option
162    OpenCodeZenGPT54Mini,
163    /// Claude Sonnet 4.6 - Anthropic-backed OpenCode Zen coding model
164    OpenCodeZenClaudeSonnet46,
165    /// GLM-5.1 - Z.AI model served through OpenCode Zen
166    OpenCodeZenGlm51,
167    /// Kimi K2.5 - Moonshot model served through OpenCode Zen
168    OpenCodeZenKimiK25,
169
170    // OpenCode Go models
171    /// GLM-5.1 - Z.AI model included with OpenCode Go
172    OpenCodeGoGlm51,
173    /// Kimi K2.5 - Moonshot model included with OpenCode Go
174    OpenCodeGoKimiK25,
175    /// MiniMax M2.5 - OpenCode Go subscription model
176    OpenCodeGoMinimaxM25,
177    /// MiniMax M2.7 - Higher-tier OpenCode Go subscription model
178    OpenCodeGoMinimaxM27,
179
180    // Qwen models
181    /// Qwen 3.7 Max - Alibaba Cloud's flagship reasoning model with 131K context
182    Qwen37Max,
183    /// Qwen 3.6 Flash - Alibaba Cloud's fast inference model with 1M context
184    Qwen36Flash,
185    /// Qwen 3.6 Plus - Alibaba Cloud's balanced model with 131K context
186    Qwen36Plus,
187    /// DeepSeek V4 Flash via Qwen Cloud API
188    QwenDeepSeekV4Flash,
189    /// DeepSeek V4 Pro via Qwen Cloud API
190    QwenDeepSeekV4Pro,
191    /// GLM-5.1 via Qwen Cloud API
192    QwenGlm51,
193
194    // Ollama models
195    /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
196    OllamaGptOss20b,
197    /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
198    OllamaGptOss20bCloud,
199    /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
200    OllamaGptOss120bCloud,
201    /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
202    OllamaQwen317b,
203    /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
204    OllamaQwen3CoderNext,
205    /// DeepSeek V3.2 Cloud - DeepSeek V3.2 reasoning deployment via Ollama Cloud
206    OllamaDeepseekV32Cloud,
207    /// DeepSeek V4 Flash Cloud - Fast inference DeepSeek V4 Flash model via Ollama Cloud
208    OllamaDeepseekV4FlashCloud,
209    /// DeepSeek V4 Pro Cloud - High-performance DeepSeek V4 Pro model via Ollama Cloud
210    OllamaDeepseekV4ProCloud,
211    /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
212    OllamaQwen3Next80bCloud,
213    /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
214    OllamaMinimaxM2Cloud,
215    /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
216    OllamaMinimaxM27Cloud,
217    /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
218    OllamaGlm5Cloud,
219    /// GLM-5.1 Cloud - Cloud-hosted GLM-5.1 model served via Ollama Cloud
220    OllamaGlm51Cloud,
221    /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
222    OllamaMinimaxM25Cloud,
223    /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
224    OllamaGemini3FlashPreviewCloud,
225    /// Kimi K2.6 Cloud - Moonshot Kimi K2.6 via Ollama Cloud
226    OllamaKimiK26Cloud,
227    /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
228    OllamaNemotron3SuperCloud,
229    /// Laguna XS.2 - Poolside's 33B MoE model (3B activated) for agentic coding via Ollama
230    OllamaLagunaXs2,
231
232    // MiniMax models
233    /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
234    MinimaxM27,
235    /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
236    MinimaxM25,
237
238    // OpenRouter models
239    /// Qwen3 32B - Dense 32B Qwen3 deployment
240    OpenRouterQwen332b,
241    /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
242    OpenRouterQwen330bA3b,
243    /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
244    OpenRouterQwen330bA3bInstruct2507,
245    /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
246    OpenRouterQwen330bA3bThinking2507,
247    /// Qwen3 14B - Lightweight Qwen3 14B model
248    OpenRouterQwen314b,
249    /// Qwen3 8B - Compact Qwen3 8B deployment
250    OpenRouterQwen38b,
251    /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
252    OpenRouterQwen3Next80bA3bInstruct,
253    /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
254    OpenRouterQwen3Next80bA3bThinking,
255    /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
256    OpenRouterQwen35Plus0215,
257    /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
258    OpenRouterQwen3Coder,
259    /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
260    OpenRouterQwen3CoderPlus,
261    /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
262    OpenRouterQwen3CoderFlash,
263    /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
264    OpenRouterQwen3Coder30bA3bInstruct,
265    /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
266    OpenRouterQwen3CoderNext,
267    /// DeepSeek V4 Pro - High-performance reasoning model via OpenRouter
268    OpenRouterDeepSeekV4Pro,
269    /// DeepSeek V4 Flash - Fast inference model via OpenRouter
270    OpenRouterDeepSeekV4Flash,
271    /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
272    OpenRouterDeepSeekR1,
273    /// DeepSeek V3.1 - DeepSeek V3.1 chat model via OpenRouter
274    OpenRouterDeepSeekChatV31,
275    /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
276    OpenRouterOpenAIGptOss120b,
277    /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
278    OpenRouterOpenAIGptOss120bFree,
279    /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
280    OpenRouterOpenAIGptOss20b,
281    /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
282    OpenRouterOpenAIGpt5,
283    /// OpenAI GPT-5.5 - OpenAI GPT-5.5 model accessed through OpenRouter
284    OpenRouterOpenAIGpt55,
285    /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
286    OpenRouterOpenAIGpt5Chat,
287
288    /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
289    OpenRouterGoogleGemini31ProPreview,
290
291    /// Claude Sonnet 4.5 - Anthropic Claude Sonnet 4.5 listing
292    OpenRouterAnthropicClaudeSonnet45,
293    /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
294    OpenRouterAnthropicClaudeSonnet46,
295    /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
296    OpenRouterAnthropicClaudeHaiku45,
297    /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
298    OpenRouterMistralaiMistralLarge2512,
299    /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
300    OpenRouterNexAgiDeepseekV31NexN1,
301    /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
302    OpenRouterStepfunStep35FlashFree,
303    /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
304    OpenRouterNvidiaNemotron3Super120bA12bFree,
305    /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
306    OpenRouterZaiGlm5,
307    /// GLM-4.7 - Z.AI GLM-4.7 model via OpenRouter
308    OpenRouterZaiGlm47,
309    /// GLM-5.1 - Z.AI GLM-5.1 next-gen foundation model via OpenRouter
310    OpenRouterZaiGlm51,
311    /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
312    OpenRouterMinimaxM25,
313    /// Kimi K2.6 - Moonshot AI's next-generation multimodal model via OpenRouter
314    OpenRouterMoonshotaiKimiK26,
315    /// Qwen3.7 Max - Alibaba's flagship Qwen3.7 model for coding and agentic workloads via OpenRouter
316    OpenRouterQwenQwen37Max,
317    /// Hy3 Preview - Tencent's high-efficiency MoE model for agentic workflows via OpenRouter
318    OpenRouterTencentHy3Preview,
319    /// Grok Build 0.1 - xAI's fast coding model for agentic software engineering via OpenRouter
320    OpenRouterXAiGrokBuild01,
321    /// MiMo-V2.5-Pro - Xiaomi's flagship agentic model for complex software engineering via OpenRouter
322    OpenRouterXiaomiMimoV25Pro,
323    /// Laguna XS.2 (free) - Poolside's efficient free coding agent model via OpenRouter
324    OpenRouterPoolsideLagunaXs2Free,
325    /// Laguna M.1 (free) - Poolside's flagship free coding agent model via OpenRouter
326    OpenRouterPoolsideLagunaM1Free,
327
328    // Poolside models
329    /// Laguna M.1 - Poolside's flagship MoE coding agent model
330    PoolsideLagunaM1,
331    /// Laguna XS.2 - Poolside's efficient MoE coding agent model
332    PoolsideLagunaXs2,
333}
vtcode_config/models/model_id.rs

vtcode_config/models/
model_id.rs
