All tasks / Code & data / Coding
Best local AI for coding
Open-weights models and tools that run entirely on your own hardware, ranked from public benchmarks. Local tools are rated by the best open model they run.
BTop ranked: DeepSeek-V4.1-Flash-1017 (DeepSeek), 1,670 on LMArena WebDev / Code Arena (overall).86 options listed · data verified 19 Oct 2026
Ranked list
| Tier | Name | Approach | Runs | Rated by | Price | Skill |
|---|---|---|---|---|---|---|
| B |  DeepSeek-V4.1-Flash-1017DeepSeek | Models | Local | 1,670 on LMArena WebDev / Code Arena (overall) | ~320 GB VRAM | — |
| B |  Kimi K3Moonshot AI | Models | Local | 1,655 on LMArena WebDev / Code Arena (overall) | ~1600 GB VRAM | — |
| B |  Ling-3.1-flashinclusionAI (Ant Group) | Models | Local | 41 on Artificial Analysis Intelligence Index | ~321 GB VRAM | — |
| B |  MiMo-V2.6-FlashXiaomi | Models | Local | 1,633 on LMArena WebDev / Code Arena (overall) | ~178 GB VRAM | — |
| B |  Hy4-previewTencent | Models | Local | 1,631 on LMArena WebDev / Code Arena (overall) | ~441 GB VRAM | — |
| B |  Qwen3.8-2.4T-A95BAlibaba (Qwen) | Models | Local | 40 on Artificial Analysis Intelligence Index | ~1370 GB VRAM | — |
| B | Qwen3.8-Flash-Next Alibaba (Qwen) | Models | Local | 1,629 on LMArena WebDev / Code Arena (overall) | ~104 GB VRAM | — |
| B |  OpenCodeAnomaly Innovations | Coding agents | Local | via DeepSeek-V4.1-Flash-1017 | Free | intermediate |
| B |  OllamaOllama | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | intermediate |
| B |  llama.cppggml-org | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  vLLMvLLM project | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  Unsloth (Studio / Desktop app)Unsloth AI | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | intermediate |
| B |  ClineCline | Editors & extensions | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | intermediate |
| B |  LiteLLMBerriAI | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | advanced |
| B |  gooseAgentic AI Foundation (AAIF); originally created by Block | Coding agents | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | intermediate |
| B |  LocalAIEttore Di Giacinto (mudler) | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B | Aider Aider (open-source project, Aider-AI) | Coding agents | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | intermediate |
| B |  exoEXO Labs | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  TextGen (text-generation-webui)oobabooga | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | intermediate |
| B |  SGLangSGLang project (LMSYS) | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  Kilo CodeKilo Code | Editors & extensions | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | intermediate |
| B | KTransformers KVCache.AI (Tsinghua MADSys) | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  TensorRT LLMNVIDIA | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  Portkey AI GatewayPortkey | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | advanced |
| B |  XinferenceXorbits | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  MLX LMApple (ml-explore) | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B |  GPUStackGPUStack | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | advanced |
| B | llama.vscode ggml-org | Editors & extensions | Local | via DeepSeek-V4.1-Flash-1017 Cloud models: tier S | Free | intermediate |
| B |  LM StudioElement Labs | Runtimes & gateways | Local | via DeepSeek-V4.1-Flash-1017 | Free | beginner |
| B |  Qwen CodeAlibaba (Qwen) | Coding agents | Local | via Kimi K3 Cloud models: tier B | Free | intermediate |
| B |  LMDeployInternLM (Shanghai AI Laboratory) | Runtimes & gateways | Local | via Kimi K3 | Free | advanced |
| B |  omp (oh-my-pi)Stencil Labs (can1357) | Coding agents | Local | via Qwen3.8-2.4T-A95B Cloud models: tier S | Free | advanced |
| C | MiMo-V2.6-Pro Xiaomi | Models | Local | 1,626 on LMArena WebDev / Code Arena (overall) | ~585 GB VRAM | — |
| C |  GLM-5.3Z.ai | Models | Local | 1,622 on LMArena WebDev / Code Arena (overall) | ~425 GB VRAM | — |
| C | GLM-5.3-Flash Z.ai | Models | Local | 1,607 on LMArena WebDev / Code Arena (overall) | ~185 GB VRAM | — |
| C | Qwen3.8-27B Alibaba (Qwen) | Models | Local | 1,592 on LMArena WebDev / Code Arena (overall) | ~17 GB VRAM | — |
| C | DeepSeek-V4-Pro-0813 DeepSeek | Models | Local | 1,582 on LMArena WebDev / Code Arena (overall) | ~920 GB VRAM | — |
| C |  Motif-3Motif Technologies | Models | Local | 34 on Artificial Analysis Intelligence Index | ~181 GB VRAM | — |
| C |  Step 5 PreviewStepFun | Models | Local | 1,565 on LMArena WebDev / Code Arena (overall) | ~346 GB VRAM | — |
| C |  K2-Horizon-375B-A23BMBZUAI Institute of Foundation Models | Models | Local | 31 on Artificial Analysis Intelligence Index | ~218 GB VRAM | — |
| C |  Muse Spark Open (816B-A42B)Meta | Models | Local | 1,524 on LMArena WebDev / Code Arena (overall) | ~470 GB VRAM | — |
| C |  Nemotron 3.5 Super VL 120B-A12BNVIDIA | Models | Local | 26 on Artificial Analysis Intelligence Index | ~72 GB VRAM | — |
| C | K2-Horizon-MoVA-36B-A4B MBZUAI Institute of Foundation Models | Models | Local | 25 on Artificial Analysis Intelligence Index | ~23 GB VRAM | — |
| C |  MiniMax-M3MiniMax | Models | Local | 1,482 on LMArena WebDev / Code Arena (overall) | ~245 GB VRAM | — |
| C |  Solar Open2 250BUpstage | Models | Local | 25 on Artificial Analysis Intelligence Index | ~144 GB VRAM | — |
| C | Kimi K2.7 Code Moonshot AI | Models | Local | 1,474 on LMArena WebDev / Code Arena (overall) | ~587 GB VRAM | — |
| C | Nemotron 3 Ultra 550B-A55B NVIDIA | Models | Local | 23 on Artificial Analysis Intelligence Index | ~321 GB VRAM | — |
| C |  A.X K2SK Telecom | Models | Local | 21 on Artificial Analysis Intelligence Index | ~396 GB VRAM | — |
| C | Nemotron 3.5 Bolt 74B-A9B NVIDIA | Models | Local | 21 on Artificial Analysis Intelligence Index | ~24 GB VRAM | — |
| C |  Beam 1.0 (501B-A23B)Reflection AI | Models | Local | 1,437 on LMArena WebDev / Code Arena (overall) | ~287 GB VRAM | — |
| C |  K-EXAONE 2.0 (750B-A37B)LG AI Research | Models | Local | 20 on Artificial Analysis Intelligence Index | ~428 GB VRAM | — |
| C |  LongCat-2.0Meituan (LongCat) | Models | Local | 19 on Artificial Analysis Intelligence Index | ~1014 GB VRAM | — |
| C |  InklingThinking Machines Lab | Models | Local | 1,414 on LMArena WebDev / Code Arena (overall) | ~557 GB VRAM | — |
| C | Inkling Small Thinking Machines Lab | Models | Local | 1,410 on LMArena WebDev / Code Arena (overall) | ~159 GB VRAM | — |
| C |  Command A+ (05-2026)Cohere | Models | Local | 13 on Artificial Analysis Intelligence Index | ~126 GB VRAM | — |
| C |  Devstral 2 (2512)Mistral AI | Models | Local | 1,197 on LMArena WebDev / Code Arena (overall) | ~72 GB VRAM | — |
| C |  Gemma 4 (31B)Google DeepMind | Models | Local | 1,368 on LMArena WebDev / Code Arena (overall) | ~19 GB VRAM | — |
| C |  gpt-oss-120bOpenAI | Models | Local | 12 on Artificial Analysis Intelligence Index | ~80 GB VRAM | — |
| C | gpt-oss-20b OpenAI | Models | Local | 9 on Artificial Analysis Intelligence Index | ~16 GB VRAM | — |
| C |  Granite 4.2 30BIBM | Models | Local | 13 on Artificial Analysis Intelligence Index | ~18 GB VRAM | — |
| C |  Laguna M.1Poolside | Models | Local | 1,350 on LMArena WebDev / Code Arena (overall) | ~130 GB VRAM | — |
| C |  Llama 4 MaverickMeta | Models | Local | 9 on Artificial Analysis Intelligence Index | ~230 GB VRAM | — |
| C | Mistral Medium 3.5 Mistral AI | Models | Local | 1,264 on LMArena WebDev / Code Arena (overall) | ~75 GB VRAM | — |
| C | Mistral Small 4 (2603) Mistral AI | Models | Local | 11 on Artificial Analysis Intelligence Index | ~69 GB VRAM | — |
| C | Muse Glimmer 30B Meta | Models | Local | 1,355 on LMArena WebDev / Code Arena (overall) | ~19 GB VRAM | — |
| C | Nemotron 3.5 Lightning 30B-A3B NVIDIA | Models | Local | 13 on Artificial Analysis Intelligence Index | ~20 GB VRAM | — |
| C | North Mini Code 1.0 Cohere | Models | Local | 10 on Artificial Analysis Intelligence Index | ~19 GB VRAM | — |
| C | Qwen3-Coder-Next Alibaba (Qwen) | Models | Local | 9 on Artificial Analysis Intelligence Index | ~47 GB VRAM | — |
| C | Ring-2.6-1T inclusionAI (Ant Group) | Models | Local | 17 on Artificial Analysis Intelligence Index | ~586 GB VRAM | — |
| C |  Trinity Large ThinkingArcee AI | Models | Local | 1,238 on LMArena WebDev / Code Arena (overall) | ~229 GB VRAM | — |
| C |  llamafileMozilla.ai | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | intermediate |
| C |  LLM (CLI)Simon Willison | Runtimes & gateways | Local | via GLM-5.3-Flash Cloud models: tier S | Free | advanced |
| C |  AIChatsigoden | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | advanced |
| C |  OsaurusOsaurus | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | intermediate |
| C |  mistral.rsEric Buehler | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | advanced |
| C |  llama-swapmostlygeek | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | advanced |
| C |  LemonadeAMD (lemonade-sdk) | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | intermediate |
| C |  RamaLamaRamaLama project (containers) | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | advanced |
| C |  TabbyAPI (ExLlamaV3)theroyallab | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | advanced |
| C |  Docker Model RunnerDocker | Runtimes & gateways | Local | via GLM-5.3-Flash | Free | intermediate |
| C |  MLC LLMMLC AI | Runtimes & gateways | Local | via Qwen3.8-27B | Free | advanced |
| C |  PiEarendil Inc. | Coding agents | Local | via DeepSeek-V4-Pro-0813 Cloud models: tier S | Free | intermediate |
| C |  OpenHandsOpenHands (All Hands AI) | Coding agents | Local | via Qwen3-Coder-Next Cloud models: tier S | Free | intermediate |
| C |  WebLLMMLC AI | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
| C | GenieX Qualcomm | Runtimes & gateways | Local | via Gemma 4 (31B) | Free | advanced |
| C |  Foundry LocalMicrosoft | Runtimes & gateways | Local | via gpt-oss-20b | Free | intermediate |
Tiers come only from public benchmarks (LMArena WebDev / Code Arena (overall), Artificial Analysis Intelligence Index); tools without a result of their own inherit the tier of the best model they offer. How we rank.