Best open-source models for coding and AI agents

Coding assistants and AI agents need reliable tool calling, long context and the stamina to keep making progress. These open models are built for that kind of work. Try two or three on your own tasks, because benchmark scores in announcements are reported by the developers themselves.

  • List updated November 11, 2026
  1. Kimi K3.1 Pro

    Moonshot tuned this release for coding and agent work, with a 1M-token context and image input. It is the heaviest option here and needs a multi-node cluster.

    • Moonshot AI
    • 2.8T (104B active)
    • 1M tokens
    • Kimi K3 License (Conditional)
    • from 1.72 TB
  2. LongCat-2.5

    Meituan's MIT-licensed 1.6T multimodal mixture of experts with 48B active parameters, built for long coding and terminal tasks. Meituan's own table reports 88.4 on Terminal Bench v2.1. It needs a data-center cluster, and deployment is documented only for NPU servers.

    • Meituan
    • 1.6T (48B active)
    • 1M tokens
    • MIT (Permissive)
    • from 1.1 TB
  3. GLM-5.3

    A text-only model aimed at coding, with 753B parameters (40B active) and a 1M context. The license is close to MIT, with a security-review condition only for giant hosting businesses.

    • Z.ai (Zhipu AI)
    • 753B (40B active)
    • 1M tokens
    • GLM-5.3 License (Conditional)
    • from 520 GB
  4. DeepSeek V4

    DeepSeek's MIT-licensed pair for long-context agents: a 1.6T Pro with 49B active parameters and a 284B Flash with 13B active, both with a 1M context. By our estimate the Flash needs about 184 GB at 4-bit.

    • DeepSeek
    • 1.6T (49B active)
    • 1M tokens
    • MIT (Permissive)
    • from 176 GB
  5. Ling-3.1-flash

    Ant Group's open text model for agents and coding: 560B total, 25B active, 1M context, MIT license. Mostly linear attention keeps long contexts affordable, but you need the specific SGLang or vLLM builds named on its model card.

    • inclusionAI (Ant Group)
    • 560B (25B active)
    • 1M tokens
    • MIT (Permissive)
    • from 386 GB
  6. MiniMax M3.1 Flash

    A compact 428B multimodal mixture of experts with 23B active parameters and a 1M context, trained to run in 4-bit. Read the license: commercial use needs a credit line and a notice, and written authorization above US$20 million a year.

    • MiniMax
    • 428B (23B active)
    • 1M tokens
    • MiniMax Community (Conditional)
    • from 277 GB
  7. Laguna M.1

    Poolside's 225B mixture of experts with 23B active parameters and a 256K context, under Apache 2.0 since June 2026, with a base checkpoint. In Poolside's own table DeepSeek-V4 Flash and Qwen3.5-397B score higher, and its newer S 2.1 reports better scores at half the size. It needs a multi-GPU node.

    • Poolside
    • 226B (23.4B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 153 GB
  8. Laguna S 2.1

    Poolside's 118B mixture of experts with 8B active parameters and a 1M context, under the permissive OpenMDW-1.1 license. Its own table shows 70.2% on Terminal-Bench 2.1 and 59.4% on SWE-Bench Pro, though Kimi K3 and Claude Fable 5 lead on terminal tasks. By our estimate the 96 GB Q4_K_M file needs about 106 GB of memory.

    • Poolside
    • 118B (8B active)
    • 1M tokens
    • OpenMDW-1.1 (Permissive)
    • from 81 GB
  9. Motif 3

    Motif Technologies' MIT-licensed 314B mixture of experts with only 13.2B active parameters and a 256K context, built for tool use and terminal work. Its own table shows 74.9 on Terminal-Bench 2.1 but weaker knowledge and math scores, and the card's commands use eight H200 or B200 GPUs.

    • Motif Technologies
    • 314B (13.2B active)
    • 256K tokens
    • MIT (Permissive)
    • from 206 GB
  10. Step 5 Preview

    StepFun's 600B mixture of experts with 27B active parameters, built for repository-level software engineering, long agent runs and finance. In StepFun's own chart it scores 67.7 on DeepSWE v1.1 at High effort, level with Kimi K3 and GLM-5.3 at Max effort. It needs an 8-GPU H200 node in FP8.

    • StepFun
    • 604B (27B active)
    • 1M tokens
    • StepFun 1.0 (Conditional)
    • from 417 GB
  11. Step 3.7 Flash

    StepFun's Apache 2.0 vision-language agent model with 198B parameters, 11B active, a 256K context and selectable reasoning levels. It scores 56.3 on SWE-Bench Pro in StepFun's own table, and a 111.5 GB GGUF runs on a 128 GB machine with StepFun's llama.cpp branch.

    • StepFun
    • 201B (11B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 120 GB
  12. Kimi K2.7 Code

    Moonshot's coding agent model built on Kimi K2.6: 1T parameters with 32B active, a 256K context, native INT4 and image and video input. Moonshot says it handles sequences of 4,000 or more tool calls, which is self-reported. The license is modified MIT.

    • Moonshot AI
    • 1.03T (32B active)
    • 256K tokens
    • Modified MIT (Conditional)
    • from 655 GB
  13. MiMo-V2.6

    Xiaomi's omnimodal agent models, trained with one large reinforcement-learning run across coding, general agents, visual tasks and cybersecurity. Hugging Face tags them MIT, but the repositories have no LICENSE file, so confirm the terms before building a product.

    • Xiaomi
    • 1.02T (42B active)
    • 1M tokens
    • MIT (Permissive)
    • from 6.49 GB
  14. Ripple-65B Preview

    Ricursive Intelligence's 65B dense model under Apache 2.0 with a 1M-token context, released as a preview on 4 November 2026. Its own table reports 80.5 on Terminal-Bench 2.1 against 80.1 for Reflection Beam, but Beam is ahead on SWE-bench Verified (81.6 to 76.5). The BF16 weights are about 130 GB.

    • Ricursive Intelligence
    • 64.8B
    • 1M tokens
    • Apache 2.0 (Permissive)
    • from 45 GB
  15. Muse Glimmer

    Meta's 30B dense model for local agents, distilled from Muse Spark and quantized to fit a 24 GB GPU, under Apache 2.0.

    • Meta
    • 29.6B
    • 128K tokens
    • Apache 2.0 (Permissive)
    • from 20 GB
  16. Nanbeige4.3-33B

    BOSS Zhipin's 33B dense agent model under Apache 2.0, with a 512K context, released on 24 October 2026. Its own table reports 80.9 on Terminal-Bench 2.1 against 73.0 for Qwen3.8-27B, while Qwen leads on SWE-Bench Pro. An INT4 build is about 23 GB, and the repositories have no LICENSE file.

    • Nanbeige LLM Lab
    • 33.1B
    • 512K tokens
    • Apache 2.0 (Permissive)
    • from 23 GB
  17. Laguna XS 2.1

    Poolside's 33B mixture of experts with only 3B active parameters and a 256K context, under OpenMDW-1.1, and the card says it runs on a Mac with 36 GB. In Poolside's own table Qwen3.6-35B-A3B scores higher on all four coding benchmarks, and the card warned about llama.cpp, vLLM and Ollama rough edges.

    • Poolside
    • 33.4B (3B active)
    • 256K tokens
    • OpenMDW-1.1 (Permissive)
    • from 22 GB
  18. Raon-1 mini preview

    KRAFTON's 28.8B mixture of experts with 4.3B active parameters under Apache 2.0, with a 262K context and built-in thinking. Its own table reports 70.8 on SWE-bench Verified against 63.2 for Qwen3.6-35B-A3B, but 47.0 against 63.3 on Terminal-Bench Lite. The card calls it a preview, and a 17.6 GB NVFP4 build fits one 32 GB GPU.

    • KRAFTON
    • 28.8B (4.3B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 18 GB
  19. Qwen3.6

    Alibaba's Apache 2.0 pair tuned for agentic coding: a 35B mixture of experts with only 3B active and a 27B dense model. By our estimate they need about 25 GB and 19 GB at 4-bit.

    • Alibaba Qwen
    • 36B (3B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 19 GB
  20. GLM-4.7-Flash

    Z.ai's MIT-licensed 30B mixture of experts with 3B active parameters, from January 2026 and still downloaded about a million times a month. Z.ai's own table shows 59.2 on SWE-bench Verified against 34.0 for gpt-oss-20b, but those rivals are older than Qwen3.5 and Qwen3.6. Official weights are BF16 only, so use a community 4-bit build, which needs about 22 GB by our estimate.

    • Z.ai (Zhipu AI)
    • 31.2B (3B active)
    • 198K tokens
    • MIT (Permissive)
    • from 22 GB
  21. Ornith 1.5

    DeepReinforce's agentic-coding models post-trained from Qwen3.5, in 9B, 35B-A3B, 122B and 397B sizes under the MIT license. All benchmark claims are self-reported. By our estimate the 9B needs about 7 GB at 4-bit.

    • DeepReinforce
    • 403B (17B active)
    • 256K tokens
    • MIT (Permissive)
    • from 6.66 GB
  22. Sage-A1.5

    Shanghai AI Laboratory's 35B-A3B agent model, built on Qwen3.6 with text and image input, a 256K context and Apache 2.0. It adds parallel sub-agents through a tool and an opt-in self-review mode. Every score comes from the lab, and the official Q4_K_M GGUF is about 22 GB.

    • Shanghai AI Laboratory
    • 36B (3B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 2.98 GB
  23. Nex-N2.6

    Nex-AGI's Apache 2.0 agent models for computer use and coding, led by the new N2.6 on GLM-5.3-Flash (321B, FP8) and the N2.5 mini, Pro and Max line up to 1.6T. Scores come from Nex-AGI's own harnesses, and N2.6 does not beat its base on every row.

    • Nex-AGI
    • 1.6T
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 24 GB
  24. Flint

    Recursive's 39.8B mixture of experts with about 2.5B active per token, trained for coding agents and released under Apache 2.0. Recursive asks for at least 48 GB of GPU or unified memory and ships 4-bit builds of 22.6 to 26.0 GB, and says it generates more than 125 tokens per second on an M4 Pro MacBook Pro.

    • Recursive
    • 39.8B (2.5B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 48 GB
  25. Mellum2.1

    JetBrains' 12B mixture of experts with 2.5B active parameters, an Apache 2.0 thinking model built to work inside code repositories as a fast sub-agent. In JetBrains' own table it scores 47.0 on SWE-bench Verified, behind Qwen3.5-9B at 50.0, and its 4-bit files are about 7 to 8 GB.

    • JetBrains
    • 12.2B (2.5B active)
    • 128K tokens
    • Apache 2.0 (Permissive)
    • from 7.7 GB
  26. Qwen3.8-27B

    A 27B dense model under Apache 2.0 with coding and agent skills that fits on one GPU, which makes it a practical model to run a local coding agent on.

    • Alibaba Qwen
    • 27.8B
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 19 GB
  27. K2 Horizon

    IFM's fully open family, with a 375B-A23B flagship, a 36B-A4B that needs about 22 GB in 4-bit, and a 7B that reports 72.4 on SWE-bench Verified in IFM's own table. In the same table Qwen3.8-27B is well ahead of the dense 32B on Terminal-Bench 2.1.

    • Institute of Foundation Models (MBZUAI)
    • 379B (23B active)
    • 512K tokens
    • Apache 2.0 (Permissive)
    • from 0.745 GB
  28. AREX-2

    BAAI's 27B dense agent built on Qwen3.8 and trained to improve its answer over many rounds, under Apache 2.0. Its own table reports 81.8 on MLE-Bench Lite and 70.7 on Frontier-CS. By our estimate it needs about 19 GB at 4-bit.

    • BAAI
    • 27.4B
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 19 GB
  29. Dory 1.0

    NVIDIA's 4B looped model: one checkpoint reruns a shared block one to four times, so you trade speed for quality when you serve it. NVIDIA reports SWE-bench Verified scores of 50.3 to 61.8 across the settings. It is about 8 GB in BF16 under OpenMDW-1.1.

    • NVIDIA
    • 4B
    • 256K tokens
    • OpenMDW-1.1 (Permissive)
    • from 2.76 GB
  30. gpt-oss

    OpenAI's Apache 2.0 reasoning models with function calling and tool use. The 117B version fits one 80 GB GPU and the 21B version runs in 16 GB.

    • OpenAI
    • 117B (5.1B active)
    • 128K tokens
    • Apache 2.0 (Permissive)
    • from 15 GB
  31. Mage-VL-Cowork

    Microsoft's 4B computer-use agent under Apache 2.0 that chooses between clicking and running shell commands. It reports 66.0% on OSWorld-Verified, the best 4B result in its own table, though a 9B model scores 68.7%. It acts for real, so run it in a sandbox.

    • Microsoft
    • 4.74B
    • 32K tokens
    • Apache 2.0 (Permissive)
    • from 3.27 GB
  32. Needle 3

    Cactus Compute's 121M Apache 2.0 model that does one job: turn a request into tool calls or structured data, offline on phones and Raspberry Pi-class boards. Accuracy drops once the tool list passes about two dozen.

    • Cactus Compute
    • 121M
    • 8K tokens
    • Apache 2.0 (Permissive)
    • from 0.0835 GB
  33. Clef

    Cloudflare's 27B and 9B decision models do not write text. They return a probability for each option you define, as a fast routing and triage step inside an agent. Apache 2.0, and the 9B has a 6.7 GB 4-bit GGUF.

    • Cloudflare
    • 27.4B
    • 64K tokens
    • Apache 2.0 (Permissive)
    • from 6.49 GB
  34. Liquid d1

    Liquid AI's 0.6B to 3B decision models answer named questions about a state in one forward pass and write no text, which suits routing and triage inside an agent. Liquid measured d1-3B at 8 ms on an RTX 4090 and 50 ms on a Jetson Orin Nano. The LFM Open License is free only below US$10 million of revenue.

    • Liquid AI
    • 3.12B
    • 128K tokens
    • LFM Open 1.0 (Conditional)
    • from 0.405 GB
  35. AutoTrust Decider

    AutoTrust's open decision models on Qwen and Gemma backbones, from 9B to 35B, with typed questions, up to 256 options and calibrated probabilities. Decider Pro 35B had about 1.65 million downloads and the vision 27B about 1.55 million in the 30 days to 11 November 2026. The labels were distilled from Typesafe's closed Jev, and the scores are AutoTrust's own.

    • AutoTrust AI Lab
    • 36B (3B active)
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 6.18 GB
  36. pplx-decider

    Perplexity's Apache 2.0 decision models: a 27B and a 4B Flash that answer several typed questions about one message or image in a single pass, with a probability for every option. Perplexity's own table gives v1.2 a Decision Index of 68.09 against 62.30 for v1.1, and the weights are 52.2 GB and 9.1 GB.

    • Perplexity
    • 26.1B
    • 256K tokens
    • Apache 2.0 (Permissive)
    • from 3.13 GB
  37. Smaug-Decisions

    Abacus.AI's Apache 2.0 decision model for audio, images and text, a Gemma 4 E4B fine-tune with a 3.45 GB phone build and a 0.7B audio sibling. Its Earshot audio benchmark is its own, and downloads are still tiny, between 95 and 190 in 30 days per repository.

    • Abacus.AI
    • 8.61B
    • 128K tokens
    • Apache 2.0 (Permissive)
    • from 0.511 GB

How we chose

  • The developer positions the model for coding, agents or tool use, and the weights are open.
  • We list design, size and license rather than benchmark rankings, which are self-reported.
  • Where a model needs special serving builds or has unclear license files, the entry says so.