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 // Moonshot models
139 /// Kimi K2.6 - Moonshot.ai's latest 1T MoE flagship (32B active, MLA, MoonViT vision)
140 MoonshotKimiK26,
141 /// Kimi K2.5 - Moonshot.ai's previous flagship reasoning model
142 MoonshotKimiK25,
143
144 // OpenCode Zen models
145 /// GPT-5.4 - OpenCode Zen default flagship model
146 OpenCodeZenGPT54,
147 /// GPT-5.4 Mini - Lower-cost OpenCode Zen GPT option
148 OpenCodeZenGPT54Mini,
149 /// Claude Sonnet 4.6 - Anthropic-backed OpenCode Zen coding model
150 OpenCodeZenClaudeSonnet46,
151 /// GLM-5.1 - Z.AI model served through OpenCode Zen
152 OpenCodeZenGlm51,
153 /// Kimi K2.5 - Moonshot model served through OpenCode Zen
154 OpenCodeZenKimiK25,
155
156 // OpenCode Go models
157 /// GLM-5.1 - Z.AI model included with OpenCode Go
158 OpenCodeGoGlm51,
159 /// Kimi K2.5 - Moonshot model included with OpenCode Go
160 OpenCodeGoKimiK25,
161 /// MiniMax M2.5 - OpenCode Go subscription model
162 OpenCodeGoMinimaxM25,
163 /// MiniMax M2.7 - Higher-tier OpenCode Go subscription model
164 OpenCodeGoMinimaxM27,
165
166 // Ollama models
167 /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
168 OllamaGptOss20b,
169 /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
170 OllamaGptOss20bCloud,
171 /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
172 OllamaGptOss120bCloud,
173 /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
174 OllamaQwen317b,
175 /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
176 OllamaQwen3CoderNext,
177 /// DeepSeek V3.2 Cloud - DeepSeek V3.2 reasoning deployment via Ollama Cloud
178 OllamaDeepseekV32Cloud,
179 /// DeepSeek V4 Flash Cloud - Fast inference DeepSeek V4 Flash model via Ollama Cloud
180 OllamaDeepseekV4FlashCloud,
181 /// DeepSeek V4 Pro Cloud - High-performance DeepSeek V4 Pro model via Ollama Cloud
182 OllamaDeepseekV4ProCloud,
183 /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
184 OllamaQwen3Next80bCloud,
185 /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
186 OllamaMinimaxM2Cloud,
187 /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
188 OllamaMinimaxM27Cloud,
189 /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
190 OllamaGlm5Cloud,
191 /// GLM-5.1 Cloud - Cloud-hosted GLM-5.1 model served via Ollama Cloud
192 OllamaGlm51Cloud,
193 /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
194 OllamaMinimaxM25Cloud,
195 /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
196 OllamaGemini3FlashPreviewCloud,
197 /// Kimi K2.6 Cloud - Moonshot Kimi K2.6 via Ollama Cloud
198 OllamaKimiK26Cloud,
199 /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
200 OllamaNemotron3SuperCloud,
201
202 // MiniMax models
203 /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
204 MinimaxM27,
205 /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
206 MinimaxM25,
207
208 // OpenRouter models
209 /// Qwen3 32B - Dense 32B Qwen3 deployment
210 OpenRouterQwen332b,
211 /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
212 OpenRouterQwen330bA3b,
213 /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
214 OpenRouterQwen330bA3bInstruct2507,
215 /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
216 OpenRouterQwen330bA3bThinking2507,
217 /// Qwen3 14B - Lightweight Qwen3 14B model
218 OpenRouterQwen314b,
219 /// Qwen3 8B - Compact Qwen3 8B deployment
220 OpenRouterQwen38b,
221 /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
222 OpenRouterQwen3Next80bA3bInstruct,
223 /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
224 OpenRouterQwen3Next80bA3bThinking,
225 /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
226 OpenRouterQwen35Plus0215,
227 /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
228 OpenRouterQwen3Coder,
229 /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
230 OpenRouterQwen3CoderPlus,
231 /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
232 OpenRouterQwen3CoderFlash,
233 /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
234 OpenRouterQwen3Coder30bA3bInstruct,
235 /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
236 OpenRouterQwen3CoderNext,
237 /// DeepSeek V4 Pro - High-performance reasoning model via OpenRouter
238 OpenRouterDeepSeekV4Pro,
239 /// DeepSeek V4 Flash - Fast inference model via OpenRouter
240 OpenRouterDeepSeekV4Flash,
241 /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
242 OpenRouterDeepSeekR1,
243 /// DeepSeek V3.1 - DeepSeek V3.1 chat model via OpenRouter
244 OpenRouterDeepSeekChatV31,
245 /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
246 OpenRouterOpenAIGptOss120b,
247 /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
248 OpenRouterOpenAIGptOss120bFree,
249 /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
250 OpenRouterOpenAIGptOss20b,
251 /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
252 OpenRouterOpenAIGpt5,
253 /// OpenAI GPT-5.5 - OpenAI GPT-5.5 model accessed through OpenRouter
254 OpenRouterOpenAIGpt55,
255 /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
256 OpenRouterOpenAIGpt5Chat,
257
258 /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
259 OpenRouterGoogleGemini31ProPreview,
260
261 /// Claude Sonnet 4.5 - Anthropic Claude Sonnet 4.5 listing
262 OpenRouterAnthropicClaudeSonnet45,
263 /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
264 OpenRouterAnthropicClaudeSonnet46,
265 /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
266 OpenRouterAnthropicClaudeHaiku45,
267 /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
268 OpenRouterMistralaiMistralLarge2512,
269 /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
270 OpenRouterNexAgiDeepseekV31NexN1,
271 /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
272 OpenRouterStepfunStep35FlashFree,
273 /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
274 OpenRouterNvidiaNemotron3Super120bA12bFree,
275 /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
276 OpenRouterZaiGlm5,
277 /// GLM-4.7 - Z.AI GLM-4.7 model via OpenRouter
278 OpenRouterZaiGlm47,
279 /// GLM-5.1 - Z.AI GLM-5.1 next-gen foundation model via OpenRouter
280 OpenRouterZaiGlm51,
281 /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
282 OpenRouterMinimaxM25,
283 /// Kimi K2.6 - Moonshot AI's next-generation multimodal model via OpenRouter
284 OpenRouterMoonshotaiKimiK26,
285}