Code Generation on MBPP+ (Score)

94.2Score

GLM-4.6

Updated 4d ago

Evaluation Results

Method	Links
GLM-4.6 2026.01		94.2
DeepSeek-V3.1 2026.01		92.6
Trinity Large Base 2026.02		88.62
A.X K1 2026.01		85.7
Qwen3-Coder-30B-A3B-Instruct (Base) 2026.02		75.13
DeepSeek V3.1 2026.02		72.2
MiMo-V2 Flash 2026.02		71.4
Step 3.5 Flash 2026.02		70.6
Qwen 3 32B 2025.12		70.6
DS-R1 32B 2025.12		70.1
DeepSeek V3.2 2026.02		69.8
Qwen 3 8B 2025.12		69.1
Olmo 3.1 Think 32B 2025.12		68.3
Olmo 3 Think (Final 3.0) 2025.12		68
Olmo 3 Think (DPO) 2025.12		67.2
Olmo 3 Think (SFT) 2025.12		66.7
Qwen 3 VL 8B Inst 2025.12		66.3
Qwen 3 VL 32B Think 2025.12		66.2
Nemotron Nano 9B v2 2025.12		66.1
K2-V2 70B Instruct 2025.12		66
Olmo 3 7B Think 2025.12		64.7
Qwen 3 8B 2025.12		64.4
DS-R1 Qwen 7B 2025.12		63.5
Olmo 3 7B Think 2025.12		63.2
Olmo 3 7B Think 2025.12		63
Qwen 3 VL 8B Think 2025.12		63
Qwen 2.5 7B 2025.12		62.6
OpenThinker3 7B 2025.12		61.4
OR Nemotron 7B 2025.12		61.2
Olmo 3 7B Instruct 2025.12		60.2
Qwen3-4B-Instruct-2507 (C/E Weighted) 2026.02		58.73
Qwen3-4B-Instruct-2507 (Base) 2026.02		56.61
Qwen3-4B-Instruct-2507 (Edge Only) 2026.02		56.61
Olmo 3 7B Instruct 2025.12		56.5
Qwen3-4B-Instruct-2507 (Basic Only) 2026.02		56.34
Olmo 3 7B Instruct 2025.12		55.9
Granite 3.3 8B Inst 2025.12		54
Qwen3-4B-Instruct-2507 (Uniform) 2026.02		53.7
Qwen3-4B-Instruct-2507 (B/I Weighted) 2026.02		52.38
Qwen3-4B-Instruct-2507 (Complex Only) 2026.02		51.85
Apertus 8B Inst 2025.12		42.1
OLMo 2 7B Inst 2025.12		40.7
Qwen3-4B-Instruct-2507 (C/E Weighted (Rev)) 2026.02		35.98