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    #[default]
32    Gemini35Flash,
33
34    // OpenAI models
35    /// GPT-5.5 - Next-gen OpenAI model dated release (2026-04-23)
36    GPT55,
37    /// GPT-5.4 - Mainline frontier GPT model for general-purpose and coding work
38    GPT54,
39    /// GPT-5.4 Pro - Higher-compute GPT-5.4 variant for difficult problems
40    GPT54Pro,
41    /// GPT-5.4 Nano - Lightweight GPT-5.4 variant optimized for speed and cost-efficiency
42    GPT54Nano,
43    /// GPT-5.4 Mini - Compact GPT-5.4 variant for cost-effective tasks with reduced reasoning overhead
44    GPT54Mini,
45    /// GPT-5.3 Codex - Code-focused GPT-5.3 variant optimized for agentic coding with reasoning effort support (low, medium, high, xhigh)
46    GPT53Codex,
47    /// GPT-OSS 20B - OpenAI's open-source 20B parameter model using harmony
48    OpenAIGptOss20b,
49    /// GPT-OSS 120B - OpenAI's open-source 120B parameter model using harmony
50    OpenAIGptOss120b,
51
52    // Anthropic models
53    /// Claude Opus 4.8 - Anthropic's most capable model for complex reasoning and agentic coding
54    ClaudeOpus48,
55    /// Claude Sonnet 4.6 - Balanced flagship Anthropic model in VT Code's conservative rollout
56    ClaudeSonnet46,
57    /// Claude Haiku 4.5 - Latest efficient Anthropic model (2025-10-15)
58    ClaudeHaiku45,
59    /// Claude Mythos Preview - Invitation-only Anthropic research preview for defensive cybersecurity workflows
60    ClaudeMythosPreview,
61    /// GitHub Copilot auto model selection
62    CopilotAuto,
63    /// GitHub Copilot GPT-5.2 Codex
64    CopilotGPT52Codex,
65    /// GitHub Copilot GPT-5.1 Codex Max
66    CopilotGPT51CodexMax,
67    /// GitHub Copilot GPT-5.4
68    CopilotGPT54,
69    /// GitHub Copilot GPT-5.4 Mini
70    CopilotGPT54Mini,
71    /// GitHub Copilot Claude Sonnet 4.6
72    CopilotClaudeSonnet46,
73
74    // DeepSeek models
75    /// DeepSeek V4 Pro - High-performance reasoning model with advanced thinking
76    DeepSeekV4Pro,
77    /// DeepSeek V4 Flash - Fast inference model for cost-effective reasoning
78    DeepSeekV4Flash,
79
80    // Mistral AI models
81    /// Mistral Large 3 - State-of-the-art open-weight general-purpose multimodal model
82    MistralLarge3,
83    // Hugging Face models
84    /// OpenAI GPT-OSS 20B via Hugging Face router
85    HuggingFaceOpenAIGptOss20b,
86    /// OpenAI GPT-OSS 120B via Hugging Face router
87    HuggingFaceOpenAIGptOss120b,
88    /// MiniMax M2.5 via Novita on Hugging Face router
89    HuggingFaceMinimaxM25Novita,
90    /// Z.AI GLM-5 via Novita on Hugging Face router
91    HuggingFaceGlm5Novita,
92    /// Z.AI GLM-5.1 via zai-org provider on Hugging Face router
93    HuggingFaceGlm51ZaiOrg,
94    /// Qwen3-Coder-Next via Novita inference provider on Hugging Face router
95    HuggingFaceQwen3CoderNextNovita,
96    /// Qwen3.5-397B-A17B via Together inference provider on Hugging Face router
97    HuggingFaceQwen35397BA17BTogether,
98    /// Kimi K2.6 via Novita on Hugging Face router
99    HuggingFaceKimiK26Novita,
100    /// DeepSeek V4 Flash via Novita on Hugging Face router
101    HuggingFaceDeepseekV4FlashNovita,
102    /// DeepSeek V4 Pro via Together on Hugging Face router
103    HuggingFaceDeepseekV4ProTogether,
104    /// Step 3.5 Flash via Hugging Face router
105    HuggingFaceStep35Flash,
106
107    // StepFun models
108    /// Step 3.7 Flash - StepFun's flagship multimodal reasoning model with tool calling
109    StepFun37Flash,
110
111    /// GLM-5 - Flagship Z.ai foundation model for complex systems
112    ZaiGlm5,
113    /// GLM-5.1 - Next-gen Z.ai foundation model with improved reasoning
114    ZaiGlm51,
115
116    // MiMo models
117    /// MiMo V2.5 Pro - Xiaomi's flagship reasoning model with 1M context
118    MiMoV25Pro,
119    /// MiMo V2.5 - Xiaomi's omni-modal model with full-modal understanding and 1M context
120    MiMoV25,
121
122    // Moonshot models
123    /// Kimi K2.6 - Moonshot.ai's latest 1T MoE flagship (32B active, MLA, MoonViT vision)
124    MoonshotKimiK26,
125
126    // OpenCode Zen models
127    /// GPT-5.4 - OpenCode Zen default flagship model
128    OpenCodeZenGPT54,
129    /// GPT-5.4 Mini - Lower-cost OpenCode Zen GPT option
130    OpenCodeZenGPT54Mini,
131    /// Claude Sonnet 4.6 - Anthropic-backed OpenCode Zen coding model
132    OpenCodeZenClaudeSonnet46,
133    /// GLM-5.1 - Z.AI model served through OpenCode Zen
134    OpenCodeZenGlm51,
135
136    // OpenCode Go models
137    /// GLM-5.1 - Z.AI model included with OpenCode Go
138    OpenCodeGoGlm51,
139    /// MiniMax M2.5 - OpenCode Go subscription model
140    OpenCodeGoMinimaxM25,
141    /// MiniMax M2.7 - Higher-tier OpenCode Go subscription model
142    OpenCodeGoMinimaxM27,
143
144    // Qwen models
145    /// Qwen 3.7 Max - Alibaba Cloud's flagship reasoning model with 131K context
146    Qwen37Max,
147    /// Qwen 3.6 Flash - Alibaba Cloud's fast inference model with 1M context
148    Qwen36Flash,
149    /// Qwen 3.6 Plus - Alibaba Cloud's balanced model with 131K context
150    Qwen36Plus,
151    /// DeepSeek V4 Flash via Qwen Cloud API
152    QwenDeepSeekV4Flash,
153    /// DeepSeek V4 Pro via Qwen Cloud API
154    QwenDeepSeekV4Pro,
155    /// GLM-5.1 via Qwen Cloud API
156    QwenGlm51,
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 V4 Flash Cloud - Fast inference DeepSeek V4 Flash model via Ollama Cloud
170    OllamaDeepseekV4FlashCloud,
171    /// DeepSeek V4 Pro Cloud - High-performance DeepSeek V4 Pro model via Ollama Cloud
172    OllamaDeepseekV4ProCloud,
173    /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
174    OllamaQwen3Next80bCloud,
175    /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
176    OllamaMinimaxM2Cloud,
177    /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
178    OllamaMinimaxM27Cloud,
179    /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
180    OllamaGlm5Cloud,
181    /// GLM-5.1 Cloud - Cloud-hosted GLM-5.1 model served via Ollama Cloud
182    OllamaGlm51Cloud,
183    /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
184    OllamaMinimaxM25Cloud,
185    /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
186    OllamaGemini3FlashPreviewCloud,
187    /// Kimi K2.6 Cloud - Moonshot Kimi K2.6 via Ollama Cloud
188    OllamaKimiK26Cloud,
189    /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
190    OllamaNemotron3SuperCloud,
191    /// Laguna XS.2 - Poolside's 33B MoE model (3B activated) for agentic coding via Ollama
192    OllamaLagunaXs2,
193
194    // llama.cpp models
195    /// Qwen 3.6 27B - Dense Qwen 3.6 local model served through llama.cpp
196    LlamaCppQwen3627b,
197    /// Qwen 3.6 35B A3B - MoE Qwen 3.6 local model served through llama.cpp
198    LlamaCppQwen3635bA3b,
199    /// Gemma 4 26B A4B - Desktop Gemma 4 MoE model served through llama.cpp
200    LlamaCppGemma426bA4b,
201    /// Gemma 4 E4B - Tiny-footprint Gemma 4 model served through llama.cpp
202    LlamaCppGemma4E4b,
203    /// GPT-OSS 20B - OpenAI open-weight model served through llama.cpp
204    LlamaCppGptOss20b,
205    /// Step 3.5 Flash - StepFun local model served through llama.cpp
206    LlamaCppStep35Flash,
207
208    // MiniMax models
209    /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
210    MinimaxM27,
211    /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
212    MinimaxM25,
213
214    // OpenRouter models
215    /// Qwen3 32B - Dense 32B Qwen3 deployment
216    OpenRouterQwen332b,
217    /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
218    OpenRouterQwen330bA3b,
219    /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
220    OpenRouterQwen330bA3bInstruct2507,
221    /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
222    OpenRouterQwen330bA3bThinking2507,
223    /// Qwen3 14B - Lightweight Qwen3 14B model
224    OpenRouterQwen314b,
225    /// Qwen3 8B - Compact Qwen3 8B deployment
226    OpenRouterQwen38b,
227    /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
228    OpenRouterQwen3Next80bA3bInstruct,
229    /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
230    OpenRouterQwen3Next80bA3bThinking,
231    /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
232    OpenRouterQwen35Plus0215,
233    /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
234    OpenRouterQwen3Coder,
235    /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
236    OpenRouterQwen3CoderPlus,
237    /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
238    OpenRouterQwen3CoderFlash,
239    /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
240    OpenRouterQwen3Coder30bA3bInstruct,
241    /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
242    OpenRouterQwen3CoderNext,
243    /// DeepSeek V4 Pro - High-performance reasoning model via OpenRouter
244    OpenRouterDeepSeekV4Pro,
245    /// DeepSeek V4 Flash - Fast inference model via OpenRouter
246    OpenRouterDeepSeekV4Flash,
247    /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
248    OpenRouterDeepSeekR1,
249    /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
250    OpenRouterOpenAIGptOss120b,
251    /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
252    OpenRouterOpenAIGptOss120bFree,
253    /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
254    OpenRouterOpenAIGptOss20b,
255    /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
256    OpenRouterOpenAIGpt5,
257    /// OpenAI GPT-5.5 - OpenAI GPT-5.5 model accessed through OpenRouter
258    OpenRouterOpenAIGpt55,
259    /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
260    OpenRouterOpenAIGpt5Chat,
261
262    /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
263    OpenRouterGoogleGemini31ProPreview,
264
265    /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
266    OpenRouterAnthropicClaudeSonnet46,
267    /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
268    OpenRouterAnthropicClaudeHaiku45,
269    /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
270    OpenRouterMistralaiMistralLarge2512,
271    /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
272    OpenRouterNexAgiDeepseekV31NexN1,
273    /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
274    OpenRouterStepfunStep35FlashFree,
275    /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
276    OpenRouterNvidiaNemotron3Super120bA12bFree,
277    /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
278    OpenRouterZaiGlm5,
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    /// Qwen3.7 Max - Alibaba's flagship Qwen3.7 model for coding and agentic workloads via OpenRouter
286    OpenRouterQwenQwen37Max,
287    /// Hy3 Preview - Tencent's high-efficiency MoE model for agentic workflows via OpenRouter
288    OpenRouterTencentHy3Preview,
289    /// Grok Build 0.1 - xAI's fast coding model for agentic software engineering via OpenRouter
290    OpenRouterXAiGrokBuild01,
291    /// MiMo-V2.5-Pro - Xiaomi's flagship agentic model for complex software engineering via OpenRouter
292    OpenRouterXiaomiMimoV25Pro,
293    /// Laguna XS.2 (free) - Poolside's efficient free coding agent model via OpenRouter
294    OpenRouterPoolsideLagunaXs2Free,
295    /// Laguna M.1 (free) - Poolside's flagship free coding agent model via OpenRouter
296    OpenRouterPoolsideLagunaM1Free,
297
298    // Poolside models
299    /// Laguna M.1 - Poolside's flagship MoE coding agent model
300    PoolsideLagunaM1,
301    /// Laguna XS.2 - Poolside's efficient MoE coding agent model
302    PoolsideLagunaXs2,
303}
vtcode_config/models/model_id.rs

vtcode_config/models/
model_id.rs
