Best Open-Weight AI Models
Open-weight models ranked by SI Score, with the exact license each weight release carries shown next to it.
How this list is ranked. Every model whose weights are published, ranked by SI Score. “Open weights” means downloadable weights; the license column shows what you may actually do with them.
- 1 Kimi K3 Moonshot AIprovisional 71.0
- 2 MiniMax-M3 MiniMaxprovisional 68.0
- 3 GLM-5.3 Z.aiprovisional 67.7
- 4 Inkling Thinking Machines Labprovisional 66.5
- 5 DeepSeek V4.1 Flash DeepSeek 66.2
- 6 GLM-5.3-Flash Z.aiprovisional 66.2
- 7 Kimi K2 Thinking Turbo Moonshot AIprovisional 66.1
- 8 GLM-5.2 Z.aiprovisional 65.7
- 9 Qwen3.5 397B-A17B Alibaba / Qwenprovisional 64.0
- 10 Kimi K2.6 Moonshot AIprovisional 64.0
- 11 Gemma 4 26B A4B IT Googleprovisional 63.4
- 12 GLM-5.1 Z.aiprovisional 63.3
- 13 DeepSeek V4 Pro DeepSeekprovisional 63.1
- 14 DeepSeek V4 Pro 0813 DeepSeekprovisional 63.0
- 15 Gemma 4 31B IT Googleprovisional 63.0
Updated Oct 9, 2026. Task lists rank on one stated measure rather than the blended SI Score. See the methodology for how pillars and confidence are computed; a dash means the source has not reported that value.