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 4.6 - Next-gen flagship Anthropic model with extended thinking
68 ClaudeOpus46,
69 /// Claude Sonnet 4.6 - Balanced flagship Anthropic model with extended thinking
70 ClaudeSonnet46,
71 /// Claude Haiku 4.5 - Latest efficient Anthropic model (2025-10-15)
72 ClaudeHaiku45,
73 /// GitHub Copilot auto model selection
74 CopilotAuto,
75 /// GitHub Copilot GPT-5.2 Codex
76 CopilotGPT52Codex,
77 /// GitHub Copilot GPT-5.1 Codex Max
78 CopilotGPT51CodexMax,
79 /// GitHub Copilot GPT-5.4
80 CopilotGPT54,
81 /// GitHub Copilot GPT-5.4 Mini
82 CopilotGPT54Mini,
83 /// GitHub Copilot Claude Sonnet 4.6
84 CopilotClaudeSonnet46,
85
86 // DeepSeek models
87 /// DeepSeek V3.2 Chat - Fast non-thinking mode
88 DeepSeekChat,
89 /// DeepSeek V3.2 Reasoner - Thinking mode with structured reasoning output
90 DeepSeekReasoner,
91 // Hugging Face models
92 /// DeepSeek V3.2 via Hugging Face router
93 HuggingFaceDeepseekV32,
94 /// OpenAI GPT-OSS 20B via Hugging Face router
95 HuggingFaceOpenAIGptOss20b,
96 /// OpenAI GPT-OSS 120B via Hugging Face router
97 HuggingFaceOpenAIGptOss120b,
98 /// DeepSeek V3.2 via Novita on Hugging Face router
99 HuggingFaceDeepseekV32Novita,
100 /// Xiaomi MiMo-V2-Flash via Novita on Hugging Face router
101 HuggingFaceXiaomiMimoV2FlashNovita,
102 /// MiniMax M2.5 via Novita on Hugging Face router
103 HuggingFaceMinimaxM25Novita,
104 /// Z.AI GLM-5 via Novita on Hugging Face router
105 HuggingFaceGlm5Novita,
106 /// Qwen3-Coder-Next via Novita inference provider on Hugging Face router
107 HuggingFaceQwen3CoderNextNovita,
108 /// Qwen3.5-397B-A17B via Together inference provider on Hugging Face router
109 HuggingFaceQwen35397BA17BTogether,
110 /// Step 3.5 Flash via Hugging Face router
111 HuggingFaceStep35Flash,
112
113 /// GLM-5 - Flagship Z.ai foundation model for complex systems
114 ZaiGlm5,
115
116 // Moonshot models
117 /// Kimi K2.5 - Moonshot.ai's flagship reasoning model
118 MoonshotKimiK25,
119
120 // Ollama models
121 /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
122 OllamaGptOss20b,
123 /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
124 OllamaGptOss20bCloud,
125 /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
126 OllamaGptOss120bCloud,
127 /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
128 OllamaQwen317b,
129 /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
130 OllamaQwen3CoderNext,
131 /// DeepSeek V3.2 Cloud - DeepSeek V3.2 reasoning deployment via Ollama Cloud
132 OllamaDeepseekV32Cloud,
133 /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
134 OllamaQwen3Next80bCloud,
135 /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
136 OllamaMinimaxM2Cloud,
137 /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
138 OllamaMinimaxM27Cloud,
139 /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
140 OllamaGlm5Cloud,
141 /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
142 OllamaMinimaxM25Cloud,
143 /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
144 OllamaGemini3FlashPreviewCloud,
145 /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
146 OllamaNemotron3SuperCloud,
147
148 // MiniMax models
149 /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
150 MinimaxM27,
151 /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
152 MinimaxM25,
153
154 // OpenRouter models
155 /// Qwen3 32B - Dense 32B Qwen3 deployment
156 OpenRouterQwen332b,
157 /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
158 OpenRouterQwen330bA3b,
159 /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
160 OpenRouterQwen330bA3bInstruct2507,
161 /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
162 OpenRouterQwen330bA3bThinking2507,
163 /// Qwen3 14B - Lightweight Qwen3 14B model
164 OpenRouterQwen314b,
165 /// Qwen3 8B - Compact Qwen3 8B deployment
166 OpenRouterQwen38b,
167 /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
168 OpenRouterQwen3Next80bA3bInstruct,
169 /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
170 OpenRouterQwen3Next80bA3bThinking,
171 /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
172 OpenRouterQwen35Plus0215,
173 /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
174 OpenRouterQwen3Coder,
175 /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
176 OpenRouterQwen3CoderPlus,
177 /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
178 OpenRouterQwen3CoderFlash,
179 /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
180 OpenRouterQwen3Coder30bA3bInstruct,
181 /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
182 OpenRouterQwen3CoderNext,
183 /// DeepSeek V3.2 Chat - Official chat model via OpenRouter
184 OpenRouterDeepseekChat,
185 /// DeepSeek V3.2 - Standard model with thinking support via OpenRouter
186 OpenRouterDeepSeekV32,
187 /// DeepSeek V3.2 Reasoner - Thinking mode via OpenRouter
188 OpenRouterDeepseekReasoner,
189 /// DeepSeek V3.2 Speciale - Enhanced reasoning model (no tool-use)
190 OpenRouterDeepSeekV32Speciale,
191 /// DeepSeek V3.2 Exp - Experimental DeepSeek V3.2 listing
192 OpenRouterDeepSeekV32Exp,
193 /// DeepSeek Chat v3.1 - Advanced DeepSeek model via OpenRouter
194 OpenRouterDeepSeekChatV31,
195 /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
196 OpenRouterDeepSeekR1,
197 /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
198 OpenRouterOpenAIGptOss120b,
199 /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
200 OpenRouterOpenAIGptOss120bFree,
201 /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
202 OpenRouterOpenAIGptOss20b,
203 /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
204 OpenRouterOpenAIGpt5,
205 /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
206 OpenRouterOpenAIGpt5Chat,
207
208 /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
209 OpenRouterGoogleGemini31ProPreview,
210
211 /// Claude Sonnet 4.5 - Anthropic Claude Sonnet 4.5 listing
212 OpenRouterAnthropicClaudeSonnet45,
213 /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
214 OpenRouterAnthropicClaudeSonnet46,
215 /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
216 OpenRouterAnthropicClaudeHaiku45,
217 /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
218 OpenRouterMistralaiMistralLarge2512,
219 /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
220 OpenRouterNexAgiDeepseekV31NexN1,
221 /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
222 OpenRouterStepfunStep35FlashFree,
223 /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
224 OpenRouterNvidiaNemotron3Super120bA12bFree,
225 /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
226 OpenRouterZaiGlm5,
227 /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
228 OpenRouterMinimaxM25,
229}