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 4.7 - Next-gen flagship Anthropic model with adaptive thinking
72 ClaudeOpus47,
73 /// Claude Opus 4.6 - Previous Anthropic flagship retained for compatibility
74 ClaudeOpus46,
75 /// Claude Sonnet 4.6 - Balanced flagship Anthropic model in VT Code's conservative rollout
76 ClaudeSonnet46,
77 /// Claude Haiku 4.5 - Latest efficient Anthropic model (2025-10-15)
78 ClaudeHaiku45,
79 /// Claude Mythos Preview - Invitation-only Anthropic research preview for defensive cybersecurity workflows
80 ClaudeMythosPreview,
81 /// GitHub Copilot auto model selection
82 CopilotAuto,
83 /// GitHub Copilot GPT-5.2 Codex
84 CopilotGPT52Codex,
85 /// GitHub Copilot GPT-5.1 Codex Max
86 CopilotGPT51CodexMax,
87 /// GitHub Copilot GPT-5.4
88 CopilotGPT54,
89 /// GitHub Copilot GPT-5.4 Mini
90 CopilotGPT54Mini,
91 /// GitHub Copilot Claude Sonnet 4.6
92 CopilotClaudeSonnet46,
93
94 // DeepSeek models
95 /// DeepSeek V4 Pro - High-performance reasoning model with advanced thinking
96 DeepSeekV4Pro,
97 /// DeepSeek V4 Flash - Fast inference model for cost-effective reasoning
98 DeepSeekV4Flash,
99
100 // Mistral AI models
101 /// Mistral Large 3 - State-of-the-art open-weight general-purpose multimodal model
102 MistralLarge3,
103 // Hugging Face models
104 /// DeepSeek V3.2 via Hugging Face router
105 HuggingFaceDeepseekV32,
106 /// OpenAI GPT-OSS 20B via Hugging Face router
107 HuggingFaceOpenAIGptOss20b,
108 /// OpenAI GPT-OSS 120B via Hugging Face router
109 HuggingFaceOpenAIGptOss120b,
110 /// DeepSeek V3.2 via Novita on Hugging Face router
111 HuggingFaceDeepseekV32Novita,
112 /// Xiaomi MiMo-V2-Flash via Novita on Hugging Face router
113 HuggingFaceXiaomiMimoV2FlashNovita,
114 /// MiniMax M2.5 via Novita on Hugging Face router
115 HuggingFaceMinimaxM25Novita,
116 /// Z.AI GLM-5 via Novita on Hugging Face router
117 HuggingFaceGlm5Novita,
118 /// Z.AI GLM-5.1 via zai-org provider on Hugging Face router
119 HuggingFaceGlm51ZaiOrg,
120 /// Qwen3-Coder-Next via Novita inference provider on Hugging Face router
121 HuggingFaceQwen3CoderNextNovita,
122 /// Qwen3.5-397B-A17B via Together inference provider on Hugging Face router
123 HuggingFaceQwen35397BA17BTogether,
124 /// Kimi K2.6 via Novita on Hugging Face router
125 HuggingFaceKimiK26Novita,
126 /// DeepSeek V4 Flash via Novita on Hugging Face router
127 HuggingFaceDeepseekV4FlashNovita,
128 /// DeepSeek V4 Pro via Together on Hugging Face router
129 HuggingFaceDeepseekV4ProTogether,
130 /// Step 3.5 Flash via Hugging Face router
131 HuggingFaceStep35Flash,
132
133 /// GLM-5 - Flagship Z.ai foundation model for complex systems
134 ZaiGlm5,
135 /// GLM-5.1 - Next-gen Z.ai foundation model with improved reasoning
136 ZaiGlm51,
137
138 // MiMo models
139 /// MiMo V2.5 Pro - Xiaomi's flagship reasoning model with 1M context
140 MiMoV25Pro,
141 /// MiMo V2.5 - Xiaomi's omni-modal model with full-modal understanding and 1M context
142 MiMoV25,
143 /// MiMo V2 Pro - Xiaomi's strong reasoning model (1M context)
144 MiMoV2Pro,
145 /// MiMo V2 Omni - Xiaomi's multimodal vision-language model with 256K context
146 MiMoV2Omni,
147 /// MiMo V2 Flash - Xiaomi's lightweight fast model for high-throughput tasks
148 MiMoV2Flash,
149
150 // Moonshot models
151 /// Kimi K2.6 - Moonshot.ai's latest 1T MoE flagship (32B active, MLA, MoonViT vision)
152 MoonshotKimiK26,
153 /// Kimi K2.5 - Moonshot.ai's previous flagship reasoning model
154 MoonshotKimiK25,
155
156 // OpenCode Zen models
157 /// GPT-5.4 - OpenCode Zen default flagship model
158 OpenCodeZenGPT54,
159 /// GPT-5.4 Mini - Lower-cost OpenCode Zen GPT option
160 OpenCodeZenGPT54Mini,
161 /// Claude Sonnet 4.6 - Anthropic-backed OpenCode Zen coding model
162 OpenCodeZenClaudeSonnet46,
163 /// GLM-5.1 - Z.AI model served through OpenCode Zen
164 OpenCodeZenGlm51,
165 /// Kimi K2.5 - Moonshot model served through OpenCode Zen
166 OpenCodeZenKimiK25,
167
168 // OpenCode Go models
169 /// GLM-5.1 - Z.AI model included with OpenCode Go
170 OpenCodeGoGlm51,
171 /// Kimi K2.5 - Moonshot model included with OpenCode Go
172 OpenCodeGoKimiK25,
173 /// MiniMax M2.5 - OpenCode Go subscription model
174 OpenCodeGoMinimaxM25,
175 /// MiniMax M2.7 - Higher-tier OpenCode Go subscription model
176 OpenCodeGoMinimaxM27,
177
178 // Ollama models
179 /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
180 OllamaGptOss20b,
181 /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
182 OllamaGptOss20bCloud,
183 /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
184 OllamaGptOss120bCloud,
185 /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
186 OllamaQwen317b,
187 /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
188 OllamaQwen3CoderNext,
189 /// DeepSeek V3.2 Cloud - DeepSeek V3.2 reasoning deployment via Ollama Cloud
190 OllamaDeepseekV32Cloud,
191 /// DeepSeek V4 Flash Cloud - Fast inference DeepSeek V4 Flash model via Ollama Cloud
192 OllamaDeepseekV4FlashCloud,
193 /// DeepSeek V4 Pro Cloud - High-performance DeepSeek V4 Pro model via Ollama Cloud
194 OllamaDeepseekV4ProCloud,
195 /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
196 OllamaQwen3Next80bCloud,
197 /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
198 OllamaMinimaxM2Cloud,
199 /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
200 OllamaMinimaxM27Cloud,
201 /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
202 OllamaGlm5Cloud,
203 /// GLM-5.1 Cloud - Cloud-hosted GLM-5.1 model served via Ollama Cloud
204 OllamaGlm51Cloud,
205 /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
206 OllamaMinimaxM25Cloud,
207 /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
208 OllamaGemini3FlashPreviewCloud,
209 /// Kimi K2.6 Cloud - Moonshot Kimi K2.6 via Ollama Cloud
210 OllamaKimiK26Cloud,
211 /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
212 OllamaNemotron3SuperCloud,
213
214 // MiniMax models
215 /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
216 MinimaxM27,
217 /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
218 MinimaxM25,
219
220 // OpenRouter models
221 /// Qwen3 32B - Dense 32B Qwen3 deployment
222 OpenRouterQwen332b,
223 /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
224 OpenRouterQwen330bA3b,
225 /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
226 OpenRouterQwen330bA3bInstruct2507,
227 /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
228 OpenRouterQwen330bA3bThinking2507,
229 /// Qwen3 14B - Lightweight Qwen3 14B model
230 OpenRouterQwen314b,
231 /// Qwen3 8B - Compact Qwen3 8B deployment
232 OpenRouterQwen38b,
233 /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
234 OpenRouterQwen3Next80bA3bInstruct,
235 /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
236 OpenRouterQwen3Next80bA3bThinking,
237 /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
238 OpenRouterQwen35Plus0215,
239 /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
240 OpenRouterQwen3Coder,
241 /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
242 OpenRouterQwen3CoderPlus,
243 /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
244 OpenRouterQwen3CoderFlash,
245 /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
246 OpenRouterQwen3Coder30bA3bInstruct,
247 /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
248 OpenRouterQwen3CoderNext,
249 /// DeepSeek V4 Pro - High-performance reasoning model via OpenRouter
250 OpenRouterDeepSeekV4Pro,
251 /// DeepSeek V4 Flash - Fast inference model via OpenRouter
252 OpenRouterDeepSeekV4Flash,
253 /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
254 OpenRouterDeepSeekR1,
255 /// DeepSeek V3.1 - DeepSeek V3.1 chat model via OpenRouter
256 OpenRouterDeepSeekChatV31,
257 /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
258 OpenRouterOpenAIGptOss120b,
259 /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
260 OpenRouterOpenAIGptOss120bFree,
261 /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
262 OpenRouterOpenAIGptOss20b,
263 /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
264 OpenRouterOpenAIGpt5,
265 /// OpenAI GPT-5.5 - OpenAI GPT-5.5 model accessed through OpenRouter
266 OpenRouterOpenAIGpt55,
267 /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
268 OpenRouterOpenAIGpt5Chat,
269
270 /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
271 OpenRouterGoogleGemini31ProPreview,
272
273 /// Claude Sonnet 4.5 - Anthropic Claude Sonnet 4.5 listing
274 OpenRouterAnthropicClaudeSonnet45,
275 /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
276 OpenRouterAnthropicClaudeSonnet46,
277 /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
278 OpenRouterAnthropicClaudeHaiku45,
279 /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
280 OpenRouterMistralaiMistralLarge2512,
281 /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
282 OpenRouterNexAgiDeepseekV31NexN1,
283 /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
284 OpenRouterStepfunStep35FlashFree,
285 /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
286 OpenRouterNvidiaNemotron3Super120bA12bFree,
287 /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
288 OpenRouterZaiGlm5,
289 /// GLM-4.7 - Z.AI GLM-4.7 model via OpenRouter
290 OpenRouterZaiGlm47,
291 /// GLM-5.1 - Z.AI GLM-5.1 next-gen foundation model via OpenRouter
292 OpenRouterZaiGlm51,
293 /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
294 OpenRouterMinimaxM25,
295 /// Kimi K2.6 - Moonshot AI's next-generation multimodal model via OpenRouter
296 OpenRouterMoonshotaiKimiK26,
297}