List your company

Reasoning on GPQA Diamond (Score %)

85.5GPQA Diamond Reasoning ScoreQwen-3.5-27B
18.83636.14353.4570.757Apr 17, 2026Apr 22, 2026Apr 27, 2026May 2, 2026May 7, 2026May 12, 2026May 17, 2026

Evaluation results

MethodLinks
Qwen-3.5-27BReasoning-mode=true2026.0585.5
Gemma-4-31B2026.0584.3
Qwen-3.5-9BReasoning-mode=true2026.0581.7
DeepSeek-R12026.0571.5
Qwen-3-32B2026.0565.2
DeepSeek-V32026.0559.1
Llama-3.3-70B2026.0550.5
Soft+Backbone=olmo-3.1-32b-it-sft2026.0449.1
Qwen-2.5-72B2026.0549
Pair+Backbone=olmo-3.1-32b-it-sft2026.0448.9
MPO+Backbone=olmo-3.1-32b-it-sft2026.0448.5
DPOBackbone=olmo-3.1-32b-it-sft2026.0447.8
Llama-3.1-70B2026.0546.7
Mag+Backbone=olmo-3.1-32b-it-sft2026.0446.6
Mistral-Small-3.2-24B2026.0546.1
GPT-4o-Mini2026.0540.2
MPO+Backbone=olmo-3-7b-it-sft2026.0437.3
Mag+Backbone=olmo-3-7b-it-sft2026.0436.9
Pair+Backbone=olmo-3-7b-it-sft2026.0436.8
Soft+Backbone=olmo-3-7b-it-sft2026.0435.4
Gemma-3-12B2026.0534.9
BaseBackbone=olmo-3.1-32b-it-sft2026.0434.1
Qwen-2.5-7B2026.0533.8
DPOBackbone=olmo-3-7b-it-sft2026.0433.7
RFTBackbone=olmo-3.1-32b-it-sft2026.0433.2
Mixtral-8x22B2026.0533.2
BaseBackbone=olmo-3-7b-it-sft2026.0432.1
Llama-3.1-8B2026.0530.4
RFTBackbone=olmo-3-7b-it-sft2026.0427.9
Soft+Backbone=gemma-3-4b-sft2026.0426.7
DPOBackbone=gemma-3-4b-sft2026.0426.3
MPO+Backbone=gemma-3-4b-sft2026.0425.5
Mag+Backbone=gemma-3-4b-sft2026.0424.1
Pair+Backbone=gemma-3-4b-sft2026.0423.6
BaseBackbone=gemma-3-4b-sft2026.0422.3
RFTBackbone=gemma-3-4b-sft2026.0421.4