Open-Weight Models in 2026

178 on record. Sorted by context; every figure comes from the OpenRouter model catalogue, with the page it was read from on each entry.

178listed
#ModelContext windowInput / 1MOutput / 1M
126Kimi K2 0711MoonshotAIopen weightstext131K$0.571$2.31
127Llama 3.1 70B InstructMetaopen weightstext131K$0.41$0.41
128Llama 3.1 8B InstructMetaopen weightstext131K$0.051$0.081
129Llama 3.1 Euryale 70B v2.2Sao10Kopen weightstext131K$0.851$0.851
130Llama 3.2 3B InstructMetaopen weightstext131K$0.051$0.331
131Llama 3.3 70B InstructMetaopen weightstext131K$0.11$0.321
132Llama 3.3 Euryale 70BSao10Kopen weightstext131K$0.651$0.751
133Ministral 3 3B 2512Mistralopen weightstext, image131K$0.11$0.11
134Mistral NemoMistralopen weightstext131K$0.019$0.031
135Muse Glimmer 30BMetaopen weightstext, image131K$0.351$1.51
136Nemotron 3.5 Content SafetyNVIDIAopen weightstext, image131K$0.21$0.21
137Qwen3 14BQwenopen weightstext131K$0.121$0.241
138Qwen3 235B A22BQwenopen weightstext131K$0.455$1.821
139Qwen3 235B A22B Thinking 2507Qwenopen weightstext131K$0.231$2.31
140Qwen3 30B A3BQwenopen weightstext131K$0.121$0.51
141Qwen3 32BQwenopen weightstext131K$0.081$0.281
142Qwen3 8BQwenopen weightstext131K$0.117$0.455
143Qwen3 VL 235B A22B ThinkingQwenopen weightstext, image131K$0.41$41
144Qwen3 VL 32B InstructQwenopen weightstext, image131K$0.104$0.416
145Qwen3 VL 8B ThinkingQwenopen weightsimage, text131K$0.181$2.11
146Granite 4.0 MicroIBMopen weightstext131K$0.017$0.112
147Mistral Small 3.1 24BMistralopen weightstext, image128K$0.351$0.555
148Nemotron 3.5 Content Safety (free)NVIDIAopen weightstext, image128Kfreefree
149Qwen2.5 VL 72B InstructQwenopen weightstext, image128K$0.81$11
150Schematron V2 SmallInference.netopen weightstext128K$0.051$0.231

Context bars share one log scale, 1K to 2M tokens. Prices are what each provider publishes per million tokens; 'varies' is a router that bills the model it picks.

More in Models

All models lists