Skip to content
LLM Pulse

Lab

huggingface

67 canonical models tracked · RSS feed for this lab

ModelBest in /MBest out /MContextProvidersReleased
DeepSeek-R1$0.7$2.501
DeepSeek-R1-0528$3$51
DeepSeek-V3$0.4$1.301
DeepSeek V3 0324$0.27$1.121
DeepSeek-V3.1$0.27$11
DeepSeek-V3.2$0.28$0.41
DeepSeek V4 Flash$0.14$0.281
DeepSeek V4 Flash 0731$0.14$0.281
DeepSeek V4 Pro$0.435$0.871
DeepSeek V4 Pro 0813$1.32$3.961
Gemma 4 26B A4B IT$0.13$0.41
Gemma 4 31B IT$0.14$0.41
Llama-3.1-8B-Instruct$0.06$0.061
Llama-3.3-70B-Instruct$0.59$0.791
MiniMax-M2$0.3$1.201
MiniMax-M2.1$0.3$1.201
MiniMax-M2.5$0.3$1.201
MiniMax-M2.7$0.3$1.201
MiniMax-M3$0.3$1.201
Kimi-K2-Instruct$1$31
Kimi-K2-Instruct-0905$1$31
Kimi-K2-Thinking$0.6$2.501
Kimi-K2.5$0.6$31
Kimi-K2.6$0.95$41
Kimi K2.7 Code$0.95$41
Kimi K3$3$151
Qwen2.5-Coder-32B-Instruct$0.06$0.21
Qwen3 235B-A22B$0.2$0.81
Qwen3 235B-A22B Instruct 2507$0.855$2.561
Qwen3-235B-A22B-Thinking-2507$0.3$31
Qwen3 30B A3B$0.12$0.51
Qwen3 32B$0.29$0.591
Qwen3-Coder 30B-A3B Instruct$0.07$0.261
Qwen3-Coder-480B-A35B-Instruct$2$21
Qwen3-Coder-Next$0.2$1.501
Qwen 3 Embedding 4B$0.01$01
Qwen 3 Embedding 8B$0.01$01
Qwen3-Next-80B-A3B-Instruct$0.25$11
Qwen3-Next-80B-A3B-Thinking$0.3$21
Qwen3 VL 235B A22B Instruct$0.3$1.501
Qwen3 VL 235B A22B Thinking$0.98$3.951
Qwen3.5 122B-A10B$0.4$3.201
Qwen3.5 27B$0.3$2.401
Qwen3.5 35B-A3B$0.25$21
Qwen3.5-397B-A17B$0.6$3.601
Qwen3.5 9B$0.17$0.251
Qwen3.6 27B$0.47$3.191
Qwen3.6 35B-A3B$0.15$0.951
Qwen3.8 2.4T A95B$2.50$6.251
Step 3.5 Flash$0.1$0.31
Step 3.7 Flash$0.2$1.151
Hy3$0.14$0.581
Inkling$1$4.051
Inkling Small$0.5$1.201
MiMo-V2-Flash$0.1$0.31
MiMo-V2.5$0.4$21
MiMo-V2.5-Pro$1$31
GLM-4.5$0.6$2.201
GLM-4.5-Air$0.13$0.851
GLM-4.5V$0.6$1.801
GLM-4.6$0.55$2.201
GLM-4.6V-Flash$0.3$0.91
GLM-4.7$0.6$2.201
GLM-4.7-Flash1
GLM-5$1$3.201
GLM-5.1$1$3.201
GLM-5.2$1.40$4.401