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