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
14/// Centralized enum for all supported model identifiers
15#[cfg_attr(feature = "schema", derive(schemars::JsonSchema))]
16#[derive(Clone, Copy, Debug, Default, PartialEq, Eq, Hash, Serialize, Deserialize)]
17pub enum ModelId {
18 // Gemini models
19 /// Gemini 3.1 Pro Preview - Latest Gemini 3.1 Pro flagship
20 Gemini31ProPreview,
21 /// Gemini 3.1 Pro Preview Custom Tools - Optimized for custom tools & bash
22 Gemini31ProPreviewCustomTools,
23 /// Gemini 3.1 Flash Lite Preview - Most cost-efficient model, offering fastest performance for high-frequency, lightweight tasks
24 Gemini31FlashLitePreview,
25 /// Gemini 3 Flash Preview - Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding
26 #[default]
27 Gemini3FlashPreview,
28
29 // OpenAI models
30 /// GPT-5 - Latest most capable OpenAI model (2025-08-07)
31 GPT5,
32 /// GPT-5.2 - Latest flagship general-purpose OpenAI model (2025-12-11)
33 GPT52,
34 /// GPT-5.2 Codex - Code-focused GPT-5.2 variant optimized for agentic coding with xhigh reasoning support
35 GPT52Codex,
36 /// GPT-5.4 - Mainline frontier GPT model for general-purpose and coding work
37 GPT54,
38 /// GPT-5.4 Pro - Higher-compute GPT-5.4 variant for difficult problems
39 GPT54Pro,
40 /// GPT-5.4 Nano - Lightweight GPT-5.4 variant optimized for speed and cost-efficiency
41 GPT54Nano,
42 /// GPT-5.4 Mini - Compact GPT-5.4 variant for cost-effective tasks with reduced reasoning overhead
43 GPT54Mini,
44 /// GPT-5.3 Codex - Code-focused GPT-5.3 variant optimized for agentic coding with reasoning effort support (low, medium, high, xhigh)
45 GPT53Codex,
46 /// GPT-5.1 Codex - Code-focused GPT-5.1 variant optimized for agentic coding
47 GPT51Codex,
48 /// GPT-5.1 Codex Max - Higher-compute GPT-5.1 Codex variant for longer-running engineering tasks
49 GPT51CodexMax,
50 /// GPT-5 Mini - Latest efficient OpenAI model (2025-08-07)
51 GPT5Mini,
52 /// GPT-5 Nano - Latest most cost-effective OpenAI model (2025-08-07)
53 GPT5Nano,
54 /// GPT-5 Codex - Code-focused GPT-5 variant optimized for agentic coding
55 GPT5Codex,
56 /// GPT-OSS 20B - OpenAI's open-source 20B parameter model using harmony
57 OpenAIGptOss20b,
58 /// GPT-OSS 120B - OpenAI's open-source 120B parameter model using harmony
59 OpenAIGptOss120b,
60
61 // Anthropic models
62 /// Claude Opus 4.6 - Next-gen flagship Anthropic model with extended thinking
63 ClaudeOpus46,
64 /// Claude Sonnet 4.6 - Balanced flagship Anthropic model with extended thinking
65 ClaudeSonnet46,
66 /// Claude Haiku 4.5 - Latest efficient Anthropic model (2025-10-15)
67 ClaudeHaiku45,
68
69 // DeepSeek models
70 /// DeepSeek V3.2 Chat - Fast non-thinking mode
71 DeepSeekChat,
72 /// DeepSeek V3.2 Reasoner - Thinking mode with structured reasoning output
73 DeepSeekReasoner,
74 // Hugging Face models
75 /// DeepSeek V3.2 via Hugging Face router
76 HuggingFaceDeepseekV32,
77 /// OpenAI GPT-OSS 20B via Hugging Face router
78 HuggingFaceOpenAIGptOss20b,
79 /// OpenAI GPT-OSS 120B via Hugging Face router
80 HuggingFaceOpenAIGptOss120b,
81 /// DeepSeek V3.2 via Novita on Hugging Face router
82 HuggingFaceDeepseekV32Novita,
83 /// Xiaomi MiMo-V2-Flash via Novita on Hugging Face router
84 HuggingFaceXiaomiMimoV2FlashNovita,
85 /// MiniMax M2.5 via Novita on Hugging Face router
86 HuggingFaceMinimaxM25Novita,
87 /// Z.AI GLM-5 via Novita on Hugging Face router
88 HuggingFaceGlm5Novita,
89 /// Qwen3-Coder-Next via Novita inference provider on Hugging Face router
90 HuggingFaceQwen3CoderNextNovita,
91 /// Qwen3.5-397B-A17B via Together inference provider on Hugging Face router
92 HuggingFaceQwen35397BA17BTogether,
93 /// Step 3.5 Flash via Hugging Face router
94 HuggingFaceStep35Flash,
95
96 /// GLM-5 - Flagship Z.ai foundation model for complex systems
97 ZaiGlm5,
98
99 // Moonshot models
100 /// Kimi K2.5 - Moonshot.ai's flagship reasoning model
101 MoonshotKimiK25,
102
103 // Ollama models
104 /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
105 OllamaGptOss20b,
106 /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
107 OllamaGptOss20bCloud,
108 /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
109 OllamaGptOss120bCloud,
110 /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
111 OllamaQwen317b,
112 /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
113 OllamaQwen3CoderNext,
114 /// DeepSeek V3.2 Cloud - DeepSeek V3.2 reasoning deployment via Ollama Cloud
115 OllamaDeepseekV32Cloud,
116 /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
117 OllamaQwen3Next80bCloud,
118 /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
119 OllamaMinimaxM2Cloud,
120 /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
121 OllamaGlm5Cloud,
122 /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
123 OllamaMinimaxM25Cloud,
124 /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
125 OllamaGemini3FlashPreviewCloud,
126 /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
127 OllamaNemotron3SuperCloud,
128
129 // MiniMax models
130 /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
131 MinimaxM27,
132 /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
133 MinimaxM25,
134
135 // OpenRouter models
136 /// Qwen3 32B - Dense 32B Qwen3 deployment
137 OpenRouterQwen332b,
138 /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
139 OpenRouterQwen330bA3b,
140 /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
141 OpenRouterQwen330bA3bInstruct2507,
142 /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
143 OpenRouterQwen330bA3bThinking2507,
144 /// Qwen3 14B - Lightweight Qwen3 14B model
145 OpenRouterQwen314b,
146 /// Qwen3 8B - Compact Qwen3 8B deployment
147 OpenRouterQwen38b,
148 /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
149 OpenRouterQwen3Next80bA3bInstruct,
150 /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
151 OpenRouterQwen3Next80bA3bThinking,
152 /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
153 OpenRouterQwen35Plus0215,
154 /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
155 OpenRouterQwen3Coder,
156 /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
157 OpenRouterQwen3CoderPlus,
158 /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
159 OpenRouterQwen3CoderFlash,
160 /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
161 OpenRouterQwen3Coder30bA3bInstruct,
162 /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
163 OpenRouterQwen3CoderNext,
164 /// DeepSeek V3.2 Chat - Official chat model via OpenRouter
165 OpenRouterDeepseekChat,
166 /// DeepSeek V3.2 - Standard model with thinking support via OpenRouter
167 OpenRouterDeepSeekV32,
168 /// DeepSeek V3.2 Reasoner - Thinking mode via OpenRouter
169 OpenRouterDeepseekReasoner,
170 /// DeepSeek V3.2 Speciale - Enhanced reasoning model (no tool-use)
171 OpenRouterDeepSeekV32Speciale,
172 /// DeepSeek V3.2 Exp - Experimental DeepSeek V3.2 listing
173 OpenRouterDeepSeekV32Exp,
174 /// DeepSeek Chat v3.1 - Advanced DeepSeek model via OpenRouter
175 OpenRouterDeepSeekChatV31,
176 /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
177 OpenRouterDeepSeekR1,
178 /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
179 OpenRouterOpenAIGptOss120b,
180 /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
181 OpenRouterOpenAIGptOss120bFree,
182 /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
183 OpenRouterOpenAIGptOss20b,
184 /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
185 OpenRouterOpenAIGpt5,
186 /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
187 OpenRouterOpenAIGpt5Chat,
188
189 /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
190 OpenRouterGoogleGemini31ProPreview,
191
192 /// Claude Sonnet 4.5 - Anthropic Claude Sonnet 4.5 listing
193 OpenRouterAnthropicClaudeSonnet45,
194 /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
195 OpenRouterAnthropicClaudeSonnet46,
196 /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
197 OpenRouterAnthropicClaudeHaiku45,
198 /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
199 OpenRouterMistralaiMistralLarge2512,
200 /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
201 OpenRouterNexAgiDeepseekV31NexN1,
202 /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
203 OpenRouterStepfunStep35FlashFree,
204 /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
205 OpenRouterNvidiaNemotron3Super120bA12bFree,
206 /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
207 OpenRouterZaiGlm5,
208 /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
209 OpenRouterMinimaxM25,
210}