Models
Open-weight models people run at home, with what each needs and what runs it.
23 models
| Model | Parameters | Context | Good at | Licence | Released |
|---|---|---|---|---|---|
| DeepSeek V4 Flash DeepSeek | 290.9B (20.4B active) | 1024k | chat, coding, agentic, reasoning | mit | 2026-04-22 |
| GLM 5.3 Flash Z.ai | 321.3B (25.4B active) | 1024k | chat, coding, agentic, reasoning, vision | mit | 2026-08-26 |
| Qwen 3.8 Flash Next Alibaba Qwen | 180B (61.6B active) | 256k | chat, coding, agentic, reasoning, vision | qwen-community-1.0 | 2026-08-24 |
| MiniMax M2.7 MiniMax | 228.7B (11B active) | 200k | coding, agentic, reasoning | other | 2026-04-09 |
| Qwen 3.8 27B Alibaba Qwen | 27.8B | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-08-05 |
| Qwen 3.5 122B-A10B Alibaba Qwen | 125.1B (10B active) | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-02-24 |
| Qwen 3.6 27B Alibaba Qwen | 27.8B | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-04-21 |
| Gemma 4 31B | 31.3B | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-03-11 |
| Nemotron 3 Super 120B-A12B NVIDIA | 123.6B (12B active) | 256k | chat, coding, agentic, reasoning | nvidia-nemotron-open-model-license | 2026-03-10 |
| gpt-oss 120B OpenAI | 116.8B (5.8B active) | 128k | chat, coding, agentic, reasoning | apache-2.0 | 2025-08-04 |
| Mistral Small 4 119B Mistral AI | 119.4B (7.1B active) | 1024k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-01-23 |
| Qwen 3.5 27B Alibaba Qwen | 27.8B | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-02-24 |
| Qwen 3.6 35B-A3B Alibaba Qwen | 36B (3B active) | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-04-15 |
| Gemma 4 26B-A4B | 25.8B (4B active) | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-03-11 |
| Qwen 3.5 35B-A3B Alibaba Qwen | 36B (3B active) | 256k | chat, coding, agentic, reasoning, vision | apache-2.0 | 2026-02-24 |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | 31.6B (3B active) | 256k | chat, coding, agentic, reasoning | openmdw-1.1 | 2026-08-01 |
| gpt-oss 20B OpenAI | 20.9B (4.2B active) | 128k | chat, coding, agentic, reasoning | apache-2.0 | 2025-08-04 |
| Qwen3 32B Alibaba Qwen | 32.8B | 40k | chat, coding, reasoning | apache-2.0 | 2025-04-27 |
| DeepSeek R1 Distill Qwen 32B DeepSeek | 32.8B | 128k | chat, reasoning | mit | 2025-01-20 |
| Qwen3 14B Alibaba Qwen | 14.8B | 40k | chat, coding, reasoning | apache-2.0 | 2025-04-27 |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | 8.2B | 128k | chat, reasoning | mit | 2025-05-29 |
| Phi-4 14B Microsoft | 14.7B | 16k | chat, reasoning | mit | 2024-12-11 |
| Qwen3 8B Alibaba Qwen | 8.2B | 40k | chat, reasoning | apache-2.0 | 2025-04-27 |