List your company

Long-context Reasoning on Long-context Benchmarks 16K (DocMath, Frames, LB-MQA, V2) (test)

64.1DocMathQwen3-30B-A3B-Thinking + SPELL
8.77223.13637.551.864Sep 28, 2025

Evaluation results

MethodLinks
Qwen3-30B-A3B-Thinking + SPELLModel Category=Reasoning Models, Input Length=16K2025.0964.166.57842.362.7
Qwen3-30B-A3B-Instruct + SPELLModel Category=Instruct Models, Input Length=16K2025.096363.175.141.560.7
Qwen3-30B-A3B-ThinkingModel Category=Reasoning Models, Input Length=16K2025.0962.964.575.739.760.7
Qwen3-30B-A3B-Thinking + RLVRModel Category=Reasoning Models, Input Length=16K2025.0962.764.77738.560.7
Qwen3-30B-A3B-Instruct + RLVRModel Category=Instruct Models, Input Length=16K2025.0962.559.971.839.858.5
Qwen2.5-32B-Instruct + SPELLModel Category=Instruct Models, Input Length=16K2025.0962.361.274.440.159.5
Qwen3-30B-A3B-InstructModel Category=Instruct Models, Input Length=16K2025.0962.355.370.536.956.3
Qwen3-4B-Thinking + SPELLModel Category=Reasoning Models, Input Length=16K2025.0961.956.671.636.856.7
Qwen2.5-32B + SPELLModel Category=Base Models, Input Length=16K2025.0961.850.262.134.252.1
R1-Distill-Qwen-14B + SPELLModel Category=Reasoning Models, Input Length=16K2025.0961.662.376.23959.8
Qwen3-4B-Thinking + RLVRModel Category=Reasoning Models, Input Length=16K2025.0960.556.671.133.855.5
Qwen2.5-32B-InstructModel Category=Instruct Models, Input Length=16K2025.096049.961.43651.8
Qwen2.5-32B-Instruct + RLVRModel Category=Instruct Models, Input Length=16K2025.0959.960.570.436.356.8
Qwen2.5-14B-Instruct + SPELLModel Category=Instruct Models, Input Length=16K2025.0959.662.172.836.857.8
R1-Distill-Qwen-14B + RLVRModel Category=Reasoning Models, Input Length=16K2025.0959.661.774.637.258.3
Qwen3-4B-ThinkingModel Category=Reasoning Models, Input Length=16K2025.0958.656.769.932.954.5
Qwen2.5-32B + RLVRModel Category=Base Models, Input Length=16K2025.0958.35059.532.850.2
R1-Distill-Qwen-14BModel Category=Reasoning Models, Input Length=16K2025.0957.759.272.436.256.4
Qwen2.5-14B + SPELLModel Category=Base Models, Input Length=16K2025.0957.652.66333.551.7
Qwen2.5-14B-InstructModel Category=Instruct Models, Input Length=16K2025.0956.351.66332.250.8
Qwen2.5-14B-Instruct + RLVRModel Category=Instruct Models, Input Length=16K2025.0956.159.67136.455.8
Qwen2.5-14B + RLVRModel Category=Base Models, Input Length=16K2025.0952.25163.332.949.9
R1-Distill-Llama-8B + SPELLModel Category=Reasoning Models, Input Length=16K2025.0948.953.468.430.250.2
Qwen2.5-32BModel Category=Base Models, Input Length=16K2025.0946.842.64933.743
Qwen2.5-7B-Instruct + SPELLModel Category=Instruct Models, Input Length=16K2025.0945.846.763.133.247.2
Qwen2.5-7B-Instruct + RLVRModel Category=Instruct Models, Input Length=16K2025.094548.759.630.145.9
R1-Distill-Llama-8B + RLVRModel Category=Reasoning Models, Input Length=16K2025.0943.451.467.83048.2
R1-Distill-Llama-8BModel Category=Reasoning Models, Input Length=16K2025.094250.366.827.946.8
Qwen2.5-7B + RLVRModel Category=Base Models, Input Length=16K2025.0941.8415030.240.8
Qwen2.5-7B + SPELLModel Category=Base Models, Input Length=16K2025.094039.250.932.340.6
Llama3.1-8B-Instruct + SPELLModel Category=Instruct Models, Input Length=16K2025.0939.248.961.628.444.5
Qwen2.5-7B-InstructModel Category=Instruct Models, Input Length=16K2025.0938.440.345.12938.2
Qwen2.5-14BModel Category=Base Models, Input Length=16K2025.093837.241.932.137.3
Llama3.1-8B-Instruct + RLVRModel Category=Instruct Models, Input Length=16K2025.0937.94558.827.542.3
Llama3.1-8B-InstructModel Category=Instruct Models, Input Length=16K2025.0933.245.652.529.140.1
Qwen2.5-7BModel Category=Base Models, Input Length=16K2025.0910.927.936.731.226.7