List your company

General Knowledge on CEval

90.4ScoreLongCat-Flash Chat
66.500872.705478.9185.1146Dec 30, 2025Jan 18, 2026Feb 7, 2026Feb 27, 2026Mar 18, 2026Apr 7, 2026Apr 27, 2026

Evaluation results

MethodLinks
LongCat-Flash ChatEvaluation Mode=Chat2025.1290.4—
LongCat-Flash Exp-ChatEvaluation Mode=Chat2025.1289.9—
GLM 4.6Evaluation Mode=Chat2025.1289.6—
DeepSeek V3.2Evaluation Mode=Chat2025.1289.6—
T3STraining Dataset=BOBA-200, Teacher Model=DeepSeek-R1, Student Model=Qwen3-8B2026.0184.40.07
T3STraining Dataset=S1K-200, Teacher Model=DeepSeek-R1, Student Model=Qwen3-8B2026.0183.80
Qwen3-8B (Base)Student Model=Qwen3-8B2026.0183.58—
T3STraining Dataset=BOBA-200, Teacher Model=QWQ, Student Model=Qwen3-8B2026.0183.580.04
T3STraining Dataset=S1K-200, Teacher Model=QWQ, Student Model=Qwen3-8B2026.0183.510.4
SFTTraining Dataset=S1K-200, Teacher Model=QWQ, Student Model=Qwen3-8B2026.0183.12-0.58
SFTTraining Dataset=S1K-200, Teacher Model=DeepSeek-R1, Student Model=Qwen3-8B2026.0182.88-0.46
SFTTraining Dataset=BOBA-200, Teacher Model=DeepSeek-R1, Student Model=Qwen3-8B2026.0182.84-0.75
SFTTraining Dataset=BOBA-200, Teacher Model=QWQ, Student Model=Qwen3-8B2026.0182.17-1.26
Full AttentionSetting=CPT setting2026.0473.66—
Hybrid-KSASetting=CPT setting2026.0472.66—
Hybrid-SWASetting=CPT setting2026.0472.36—
KSASetting=CPT setting2026.0472.14—
Hybrid-SCASetting=CPT setting2026.0471.66—
Hybrid-LinearSetting=CPT setting2026.0467.42—