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    /// Z.AI GLM-5.1 via DeepInfra inference provider on Hugging Face router
107    HuggingFaceGlm51Deepinfra,
108    /// MiniMax M2.7 via Novita on Hugging Face router
109    HuggingFaceMinimaxM27Novita,
110    /// DeepSeek V4 Pro via Novita on Hugging Face router
111    HuggingFaceDeepseekV4ProNovita,
112
113    // StepFun models
114    /// Step 3.7 Flash - StepFun's flagship multimodal reasoning model with tool calling
115    StepFun37Flash,
116
117    /// GLM-5 - Flagship Z.ai foundation model for complex systems
118    ZaiGlm5,
119    /// GLM-5.1 - Next-gen Z.ai foundation model with improved reasoning
120    ZaiGlm51,
121
122    // MiMo models
123    /// MiMo V2.5 Pro - Xiaomi's flagship reasoning model with 1M context
124    MiMoV25Pro,
125    /// MiMo V2.5 - Xiaomi's omni-modal model with full-modal understanding and 1M context
126    MiMoV25,
127
128    // Moonshot models
129    /// Kimi K2.6 - Moonshot.ai's latest 1T MoE flagship (32B active, MLA, MoonViT vision)
130    MoonshotKimiK26,
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
142    // OpenCode Go models
143    /// GLM-5.1 - Z.AI model included with OpenCode Go
144    OpenCodeGoGlm51,
145    /// MiniMax M2.5 - OpenCode Go subscription model
146    OpenCodeGoMinimaxM25,
147    /// MiniMax M2.7 - Higher-tier OpenCode Go subscription model
148    OpenCodeGoMinimaxM27,
149
150    // Qwen models
151    /// Qwen 3.7 Max - Alibaba Cloud's flagship reasoning model with 131K context
152    Qwen37Max,
153    /// Qwen 3.6 Flash - Alibaba Cloud's fast inference model with 1M context
154    Qwen36Flash,
155    /// Qwen 3.6 Plus - Alibaba Cloud's balanced model with 131K context
156    Qwen36Plus,
157    /// DeepSeek V4 Flash via Qwen Cloud API
158    QwenDeepSeekV4Flash,
159    /// DeepSeek V4 Pro via Qwen Cloud API
160    QwenDeepSeekV4Pro,
161    /// GLM-5.1 via Qwen Cloud API
162    QwenGlm51,
163
164    // Ollama models
165    /// GPT-OSS 20B - Open-weight GPT-OSS 20B model served via Ollama locally
166    OllamaGptOss20b,
167    /// GPT-OSS 20B Cloud - Cloud-hosted GPT-OSS 20B served via Ollama Cloud
168    OllamaGptOss20bCloud,
169    /// GPT-OSS 120B Cloud - Cloud-hosted GPT-OSS 120B served via Ollama Cloud
170    OllamaGptOss120bCloud,
171    /// Qwen3 1.7B - Qwen3 1.7B model served via Ollama
172    OllamaQwen317b,
173    /// Qwen3 Coder Next - Coding-optimized Qwen3 Next model served via Ollama locally
174    OllamaQwen3CoderNext,
175    /// DeepSeek V4 Flash Cloud - Fast inference DeepSeek V4 Flash model via Ollama Cloud
176    OllamaDeepseekV4FlashCloud,
177    /// DeepSeek V4 Pro Cloud - High-performance DeepSeek V4 Pro model via Ollama Cloud
178    OllamaDeepseekV4ProCloud,
179    /// Qwen3 Next 80B Cloud - Next-generation Qwen3 80B via Ollama Cloud
180    OllamaQwen3Next80bCloud,
181    /// MiniMax-M2 Cloud - Cloud-hosted MiniMax-M2 model served via Ollama Cloud
182    OllamaMinimaxM2Cloud,
183    /// MiniMax-M2.7 Cloud - Cloud-hosted MiniMax-M2.7 model served via Ollama Cloud
184    OllamaMinimaxM27Cloud,
185    /// GLM-5 Cloud - Cloud-hosted GLM-5 model served via Ollama Cloud
186    OllamaGlm5Cloud,
187    /// GLM-5.1 Cloud - Cloud-hosted GLM-5.1 model served via Ollama Cloud
188    OllamaGlm51Cloud,
189    /// MiniMax-M2.5 Cloud - Cloud-hosted MiniMax-M2.5 model served via Ollama Cloud
190    OllamaMinimaxM25Cloud,
191    /// Gemini 3 Flash Preview Cloud - Google Gemini 3 Flash Preview via Ollama Cloud
192    OllamaGemini3FlashPreviewCloud,
193    /// Kimi K2.6 Cloud - Moonshot Kimi K2.6 via Ollama Cloud
194    OllamaKimiK26Cloud,
195    /// Nemotron 3 Super Cloud - NVIDIA Nemotron 3 Super 120B via Ollama Cloud
196    OllamaNemotron3SuperCloud,
197    /// Laguna XS.2 - Poolside's 33B MoE model (3B activated) for agentic coding via Ollama
198    OllamaLagunaXs2,
199
200    // llama.cpp models
201    /// Qwen 3.6 27B - Dense Qwen 3.6 local model served through llama.cpp
202    LlamaCppQwen3627b,
203    /// Qwen 3.6 35B A3B - MoE Qwen 3.6 local model served through llama.cpp
204    LlamaCppQwen3635bA3b,
205    /// Gemma 4 26B A4B - Desktop Gemma 4 MoE model served through llama.cpp
206    LlamaCppGemma426bA4b,
207    /// Gemma 4 E4B - Tiny-footprint Gemma 4 model served through llama.cpp
208    LlamaCppGemma4E4b,
209    /// GPT-OSS 20B - OpenAI open-weight model served through llama.cpp
210    LlamaCppGptOss20b,
211    /// Step 3.5 Flash - StepFun local model served through llama.cpp
212    LlamaCppStep35Flash,
213
214    // MiniMax models
215    /// MiniMax-M2.7 - Recursive self-improvement flagship with 204.8K context
216    MinimaxM27,
217    /// MiniMax-M2.5 - Latest MiniMax model with further improvements in reasoning and coding
218    MinimaxM25,
219
220    // OpenRouter models
221    /// Qwen3 32B - Dense 32B Qwen3 deployment
222    OpenRouterQwen332b,
223    /// Qwen3 30B A3B - Active-parameter 30B Qwen3 model
224    OpenRouterQwen330bA3b,
225    /// Qwen3 30B A3B Instruct 2507 - Instruction-tuned Qwen3 30B A3B
226    OpenRouterQwen330bA3bInstruct2507,
227    /// Qwen3 30B A3B Thinking 2507 - Deliberative Qwen3 30B A3B release
228    OpenRouterQwen330bA3bThinking2507,
229    /// Qwen3 14B - Lightweight Qwen3 14B model
230    OpenRouterQwen314b,
231    /// Qwen3 8B - Compact Qwen3 8B deployment
232    OpenRouterQwen38b,
233    /// Qwen3 Next 80B A3B Instruct - Next-generation Qwen3 instruction model
234    OpenRouterQwen3Next80bA3bInstruct,
235    /// Qwen3 Next 80B A3B Thinking - Next-generation Qwen3 reasoning release
236    OpenRouterQwen3Next80bA3bThinking,
237    /// Qwen3.5-397B-A17B - Native vision-language model with linear attention and sparse MoE, 1M context window
238    OpenRouterQwen35Plus0215,
239    /// Qwen3 Coder - Qwen3-based coding model tuned for IDE workflows
240    OpenRouterQwen3Coder,
241    /// Qwen3 Coder Plus - Premium Qwen3 coding model with long context
242    OpenRouterQwen3CoderPlus,
243    /// Qwen3 Coder Flash - Latency optimised Qwen3 coding model
244    OpenRouterQwen3CoderFlash,
245    /// Qwen3 Coder 30B A3B Instruct - Large Mixture-of-Experts coding deployment
246    OpenRouterQwen3Coder30bA3bInstruct,
247    /// Qwen3 Coder Next - Next-generation Qwen3 coding model with enhanced reasoning
248    OpenRouterQwen3CoderNext,
249    /// DeepSeek V4 Pro - High-performance reasoning model via OpenRouter
250    OpenRouterDeepSeekV4Pro,
251    /// DeepSeek V4 Flash - Fast inference model via OpenRouter
252    OpenRouterDeepSeekV4Flash,
253    /// DeepSeek R1 - DeepSeek R1 reasoning model with chain-of-thought
254    OpenRouterDeepSeekR1,
255    /// OpenAI gpt-oss-120b - Open-weight 120B reasoning model via OpenRouter
256    OpenRouterOpenAIGptOss120b,
257    /// OpenAI gpt-oss-120b:free - Open-weight 120B reasoning model free tier via OpenRouter
258    OpenRouterOpenAIGptOss120bFree,
259    /// OpenAI gpt-oss-20b - Open-weight 20B deployment via OpenRouter
260    OpenRouterOpenAIGptOss20b,
261    /// OpenAI GPT-5 - OpenAI GPT-5 model accessed through OpenRouter
262    OpenRouterOpenAIGpt5,
263    /// OpenAI GPT-5.5 - OpenAI GPT-5.5 model accessed through OpenRouter
264    OpenRouterOpenAIGpt55,
265    /// OpenAI GPT-5 Chat - Chat optimised GPT-5 endpoint without tool use
266    OpenRouterOpenAIGpt5Chat,
267
268    /// Gemini 3.1 Pro Preview - Google's latest Gemini 3.1 Pro model via OpenRouter
269    OpenRouterGoogleGemini31ProPreview,
270
271    /// Claude Sonnet 4.6 - Anthropic Claude Sonnet 4.6 listing
272    OpenRouterAnthropicClaudeSonnet46,
273    /// Claude Haiku 4.5 - Anthropic Claude Haiku 4.5 listing
274    OpenRouterAnthropicClaudeHaiku45,
275    /// Mistral Large 3 2512 - Mistral Large 3 2512 model via OpenRouter
276    OpenRouterMistralaiMistralLarge2512,
277    /// DeepSeek V3.1 Nex N1 - Nex AGI DeepSeek V3.1 Nex N1 model via OpenRouter
278    OpenRouterNexAgiDeepseekV31NexN1,
279    /// Step 3.5 Flash (free) - StepFun's most capable open-source reasoning model via OpenRouter
280    OpenRouterStepfunStep35FlashFree,
281    /// Nemotron 3 Super (free) - NVIDIA's 120B hybrid MoE model via OpenRouter
282    OpenRouterNvidiaNemotron3Super120bA12bFree,
283    /// GLM-5 - Z.AI GLM-5 flagship foundation model via OpenRouter
284    OpenRouterZaiGlm5,
285    /// GLM-5.1 - Z.AI GLM-5.1 next-gen foundation model via OpenRouter
286    OpenRouterZaiGlm51,
287    /// MiniMax-M2.5 - MiniMax flagship model via OpenRouter
288    OpenRouterMinimaxM25,
289    /// Kimi K2.6 - Moonshot AI's next-generation multimodal model via OpenRouter
290    OpenRouterMoonshotaiKimiK26,
291    /// Qwen3.7 Max - Alibaba's flagship Qwen3.7 model for coding and agentic workloads via OpenRouter
292    OpenRouterQwenQwen37Max,
293    /// Hy3 Preview - Tencent's high-efficiency MoE model for agentic workflows via OpenRouter
294    OpenRouterTencentHy3Preview,
295    /// Grok Build 0.1 - xAI's fast coding model for agentic software engineering via OpenRouter
296    OpenRouterXAiGrokBuild01,
297    /// MiMo-V2.5-Pro - Xiaomi's flagship agentic model for complex software engineering via OpenRouter
298    OpenRouterXiaomiMimoV25Pro,
299    /// Laguna XS.2 (free) - Poolside's efficient free coding agent model via OpenRouter
300    OpenRouterPoolsideLagunaXs2Free,
301    /// Laguna M.1 (free) - Poolside's flagship free coding agent model via OpenRouter
302    OpenRouterPoolsideLagunaM1Free,
303
304    // Poolside models
305    /// Laguna M.1 - Poolside's flagship MoE coding agent model
306    PoolsideLagunaM1,
307    /// Laguna XS.2 - Poolside's efficient MoE coding agent model
308    PoolsideLagunaXs2,
309}
vtcode_config/models/model_id.rs

vtcode_config/models/
model_id.rs
