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 Flash Preview - Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding
31 #[default]
32 Gemini3FlashPreview,
33
34 // OpenAI models
35 /// GPT-5 - Latest most capable OpenAI model (2025-08-07)
36 GPT5,
37 /// GPT-5.2 - Latest flagship general-purpose OpenAI model (2025-12-11)
38 GPT52,
39 /// GPT-5.2 Codex - Code-focused GPT-5.2 variant optimized for agentic coding with xhigh reasoning support
40 GPT52Codex,
41 /// GPT-5.4 - Mainline frontier GPT model for general-purpose and coding work
42 GPT54,
43 /// GPT-5.4 Pro - Higher-compute GPT-5.4 variant for difficult problems
44 GPT54Pro,
45 /// GPT-5.4 Nano - Lightweight GPT-5.4 variant optimized for speed and cost-efficiency
46 GPT54Nano,
47 /// GPT-5.4 Mini - Compact GPT-5.4 variant for cost-effective tasks with reduced reasoning overhead
48 GPT54Mini,
49 /// GPT-5.3 Codex - Code-focused GPT-5.3 variant optimized for agentic coding with reasoning effort support (low, medium, high, xhigh)
50 GPT53Codex,
51 /// GPT-5.1 Codex - Code-focused GPT-5.1 variant optimized for agentic coding
52 GPT51Codex,
53 /// GPT-5.1 Codex Max - Higher-compute GPT-5.1 Codex variant for longer-running engineering tasks
54 GPT51CodexMax,
55 /// GPT-5 Mini - Latest efficient OpenAI model (2025-08-07)
56 GPT5Mini,
57 /// GPT-5 Nano - Latest most cost-effective OpenAI model (2025-08-07)
58 GPT5Nano,
59 /// GPT-5 Codex - Code-focused GPT-5 variant optimized for agentic coding
60 GPT5Codex,
61 /// GPT-OSS 20B - OpenAI's open-source 20B parameter model using harmony
62 OpenAIGptOss20b,
63 /// GPT-OSS 120B - OpenAI's open-source 120B parameter model using harmony
64 OpenAIGptOss120b,
65
66 // Anthropic models
67 /// Claude Opus 4.7 - Next-gen flagship Anthropic model with adaptive thinking
68 ClaudeOpus47,
69 /// Claude Opus 4.6 - Previous Anthropic flagship retained for compatibility
70 ClaudeOpus46,
71 /// Claude Sonnet 4.6 - Balanced flagship Anthropic model in VT Code's conservative rollout
72 ClaudeSonnet46,
73 /// Claude Haiku 4.5 - Latest efficient Anthropic model (2025-10-15)
74 ClaudeHaiku45,
75 /// Claude Mythos Preview - Invitation-only Anthropic research preview for defensive cybersecurity workflows
76 ClaudeMythosPreview,
77 /// GitHub Copilot auto model selection
78 CopilotAuto,
79 /// GitHub Copilot GPT-5.2 Codex
80 CopilotGPT52Codex,
81 /// GitHub Copilot GPT-5.1 Codex Max
82 CopilotGPT51CodexMax,
83 /// GitHub Copilot GPT-5.4
84 CopilotGPT54,
85 /// GitHub Copilot GPT-5.4 Mini
86 CopilotGPT54Mini,
87 /// GitHub Copilot Claude Sonnet 4.6
88 CopilotClaudeSonnet46,
89
90 // DeepSeek models
91 /// DeepSeek V3.2 Chat - Fast non-thinking mode
92 DeepSeekChat,
93 /// DeepSeek V3.2 Reasoner - Thinking mode with structured reasoning output
94 DeepSeekReasoner,
95 // Hugging Face models
96 /// DeepSeek V3.2 via Hugging Face router
97 HuggingFaceDeepseekV32,
98 /// OpenAI GPT-OSS 20B via Hugging Face router
99 HuggingFaceOpenAIGptOss20b,
100 /// OpenAI GPT-OSS 120B via Hugging Face router
101 HuggingFaceOpenAIGptOss120b,
102 /// DeepSeek V3.2 via Novita on Hugging Face router
103 HuggingFaceDeepseekV32Novita,
104 /// Xiaomi MiMo-V2-Flash via Novita on Hugging Face router
105 HuggingFaceXiaomiMimoV2FlashNovita,
106 /// MiniMax M2.5 via Novita on Hugging Face router
107 HuggingFaceMinimaxM25Novita,
108 /// Z.AI GLM-5 via Novita on Hugging Face router
109 HuggingFaceGlm5Novita,
110 /// Z.AI GLM-5.1 via zai-org provider on Hugging Face router
111 HuggingFaceGlm51ZaiOrg,
112 /// Qwen3-Coder-Next via Novita inference provider on Hugging Face router
113 HuggingFaceQwen3CoderNextNovita,
114 /// Qwen3.5-397B-A17B via Together inference provider on Hugging Face router
115 HuggingFaceQwen35397BA17BTogether,
116 /// Kimi K2.6 via Novita on Hugging Face router
117 HuggingFaceKimiK26Novita,
118 /// Step 3.5 Flash via Hugging Face router
119 HuggingFaceStep35Flash,
120
121 /// GLM-5 - Flagship Z.ai foundation model for complex systems
122 ZaiGlm5,
123 /// GLM-5.1 - Next-gen Z.ai foundation model with improved reasoning
124 ZaiGlm51,
125
126 // Moonshot models
127 /// Kimi K2.6 - Moonshot.ai's latest 1T MoE flagship (32B active, MLA, MoonViT vision)
128 MoonshotKimiK26,
129 /// Kimi K2.5 - Moonshot.ai's previous flagship reasoning model
130 MoonshotKimiK25,
131
132 // OpenCode Zen models
133 /// GPT-5.4 - OpenCode Zen default flagship model
134 OpenCodeZenGPT54,
135 /// GPT-5.4 Mini - Lower-cost OpenCode Zen GPT option
136 OpenCodeZenGPT54Mini,
137 /// Claude Sonnet 4.6 - Anthropic-backed OpenCode Zen coding model
138 OpenCodeZenClaudeSonnet46,
139 /// GLM-5.1 - Z.AI model served through OpenCode Zen
140 OpenCodeZenGlm51,
141 /// Kimi K2.5 - Moonshot model served through OpenCode Zen
142 OpenCodeZenKimiK25,
143
144 // OpenCode Go models
145 /// GLM-5.1 - Z.AI model included with OpenCode Go
146 OpenCodeGoGlm51,
147 /// Kimi K2.5 - Moonshot model included with OpenCode Go
148 OpenCodeGoKimiK25,
149 /// MiniMax M2.5 - OpenCode Go subscription model
150 OpenCodeGoMinimaxM25,
151 /// MiniMax M2.7 - Higher-tier OpenCode Go subscription model
152 OpenCodeGoMinimaxM27,
153
154 // Ollama models
155 /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
156 OllamaGptOss20b,
157 /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
158 OllamaGptOss20bCloud,
159 /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
160 OllamaGptOss120bCloud,
161 /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
162 OllamaQwen317b,
163 /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
164 OllamaQwen3CoderNext,
165 /// DeepSeek V3.2 Cloud - DeepSeek V3.2 reasoning deployment via Ollama Cloud
166 OllamaDeepseekV32Cloud,
167 /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
168 OllamaQwen3Next80bCloud,
169 /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
170 OllamaMinimaxM2Cloud,
171 /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
172 OllamaMinimaxM27Cloud,
173 /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
174 OllamaGlm5Cloud,
175 /// GLM-5.1 Cloud - Cloud-hosted GLM-5.1 model served via Ollama Cloud
176 OllamaGlm51Cloud,
177 /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
178 OllamaMinimaxM25Cloud,
179 /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
180 OllamaGemini3FlashPreviewCloud,
181 /// Kimi K2.6 Cloud - Moonshot Kimi K2.6 via Ollama Cloud
182 OllamaKimiK26Cloud,
183 /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
184 OllamaNemotron3SuperCloud,
185
186 // MiniMax models
187 /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
188 MinimaxM27,
189 /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
190 MinimaxM25,
191
192 // OpenRouter models
193 /// Qwen3 32B - Dense 32B Qwen3 deployment
194 OpenRouterQwen332b,
195 /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
196 OpenRouterQwen330bA3b,
197 /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
198 OpenRouterQwen330bA3bInstruct2507,
199 /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
200 OpenRouterQwen330bA3bThinking2507,
201 /// Qwen3 14B - Lightweight Qwen3 14B model
202 OpenRouterQwen314b,
203 /// Qwen3 8B - Compact Qwen3 8B deployment
204 OpenRouterQwen38b,
205 /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
206 OpenRouterQwen3Next80bA3bInstruct,
207 /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
208 OpenRouterQwen3Next80bA3bThinking,
209 /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
210 OpenRouterQwen35Plus0215,
211 /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
212 OpenRouterQwen3Coder,
213 /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
214 OpenRouterQwen3CoderPlus,
215 /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
216 OpenRouterQwen3CoderFlash,
217 /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
218 OpenRouterQwen3Coder30bA3bInstruct,
219 /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
220 OpenRouterQwen3CoderNext,
221 /// DeepSeek V3.2 Chat - Official chat model via OpenRouter
222 OpenRouterDeepseekChat,
223 /// DeepSeek V3.2 - Standard model with thinking support via OpenRouter
224 OpenRouterDeepSeekV32,
225 /// DeepSeek V3.2 Reasoner - Thinking mode via OpenRouter
226 OpenRouterDeepseekReasoner,
227 /// DeepSeek V3.2 Speciale - Enhanced reasoning model (no tool-use)
228 OpenRouterDeepSeekV32Speciale,
229 /// DeepSeek V3.2 Exp - Experimental DeepSeek V3.2 listing
230 OpenRouterDeepSeekV32Exp,
231 /// DeepSeek Chat v3.1 - Advanced DeepSeek model via OpenRouter
232 OpenRouterDeepSeekChatV31,
233 /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
234 OpenRouterDeepSeekR1,
235 /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
236 OpenRouterOpenAIGptOss120b,
237 /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
238 OpenRouterOpenAIGptOss120bFree,
239 /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
240 OpenRouterOpenAIGptOss20b,
241 /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
242 OpenRouterOpenAIGpt5,
243 /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
244 OpenRouterOpenAIGpt5Chat,
245
246 /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
247 OpenRouterGoogleGemini31ProPreview,
248
249 /// Claude Sonnet 4.5 - Anthropic Claude Sonnet 4.5 listing
250 OpenRouterAnthropicClaudeSonnet45,
251 /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
252 OpenRouterAnthropicClaudeSonnet46,
253 /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
254 OpenRouterAnthropicClaudeHaiku45,
255 /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
256 OpenRouterMistralaiMistralLarge2512,
257 /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
258 OpenRouterNexAgiDeepseekV31NexN1,
259 /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
260 OpenRouterStepfunStep35FlashFree,
261 /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
262 OpenRouterNvidiaNemotron3Super120bA12bFree,
263 /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
264 OpenRouterZaiGlm5,
265 /// GLM-4.7 - Z.AI GLM-4.7 model via OpenRouter
266 OpenRouterZaiGlm47,
267 /// GLM-5.1 - Z.AI GLM-5.1 next-gen foundation model via OpenRouter
268 OpenRouterZaiGlm51,
269 /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
270 OpenRouterMinimaxM25,
271 /// Kimi K2.6 - Moonshot AI's next-generation multimodal model via OpenRouter
272 OpenRouterMoonshotaiKimiK26,
273}