Share your thoughts, 1 month free Claude Pro on usSee more

Synthetic Long-context Reasoning on GraphWalks

68.91Accuracy

Qwen3-30B-A3B-Thinking-2507 + DocQA-RL-1.6K

Updated 1mo ago

Evaluation Results

Method	Links
Qwen3-30B-A3B-Thinking-2507 + DocQA-RL-1.6K 2026.06		68.91
Qwen3-30B-A3B-Thinking-2507 + Data Recipe for Long-Context RL 2026.06		67.75
Qwen3-30B-A3B-Thinking-2507 2026.06		66
Qwen3-30B-A3B-Thinking-2507 + KeyChain-15K 2026.06		65.82
Qwen3-4B-Thinking-2507 + Data Recipe for Long-Context RL 2026.06		49.74
Qwen3-4B-Thinking-2507 + DocQA-RL-1.6K 2026.06		49.42
Qwen3-4B-Thinking-2507 + KeyChain-15K 2026.06		48.7
Qwen3-4B-Thinking-2507 2026.06		48.67
Qwen3-8B-128K 2026.06		42.27
Qwen3-8B-128K + KeyChain-15K 2026.06		41.36
Qwen3-8B-128K + Data Recipe for Long-Context RL 2026.06		40.96
Qwen3-8B-128K + DocQA-RL-1.6K 2026.06		39.29