Leaderboard

Real benchmarks from real hardware. Compare local LLM performance across different setups. Run the command below to contribute yours.

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest

Trending

Last 7 days
Top Models
  • 1
    qwen3.5:4b 4
  • 2
    hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ1_S 3
  • 3
    maternion/ling-3.0-tiny:8b 3
  • 4
    qwen3.5:4b-mlx 3
  • 5
    gemma4:e4b 2
Top Machines
  • 1
    MacBook Pro 47
  • 2
    GMKtec NucBox EVO-T2S 6
  • 3
    LENOVO 20U30029MB 5
  • 4
    Micro-Star International Co., Ltd. MS-7E51 3
  • 5
    Mac mini 2
Top CPUs
  • 1
    Apple M1 Pro 46
  • 2
    Intel Core™ Ultra X7 358H 6
  • 3
    Intel Core™ i5-10210U 5
  • 4
    AMD Ryzen 7 9700X 8-Core Processor 3
  • 5
    Apple M2 2

Benchmarks

500

total runs

Models

279

unique models

Families

44

model families

Hardware

49

unique CPUs

Get notified when new models & hardware are benchmarked
Filters

500 results

#CPUModelSizeRAMRuntimetok/sTTFTHW FitQualityGlobalFlagsVerdict
1Apple M2 Maxqwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx30B/MXFP496 GBLM-STUDIOMLX79.61.7 s
95
98
97
THINKExcellent
2Apple M4 Maxqwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx30B/MXFP4128 GBLM-STUDIOMLX112.81.5 s
95
96
96
THINKExcellent
3AMD RYZEN AI MAX+ 395nemotron3:33b33.0B/Q4_K_M125 GBOLLAMAGGUF62.71.0 s
95
97
96
ECOTHINKExcellent
4Apple M4 Progpt-oss-20b64 GBLM-STUDIOGGUF118.22.2 s
94
95
95
THINKExcellent
5Apple M2 Maxqwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx30B/MXFP496 GBLM-STUDIOMLX80.41.8 s
94
96
95
THINKExcellent
6Intel Core™ Ultra 9 285Kgpt-oss:20b20.9B/MXFP447 GBOLLAMAGGUF239.9616 ms
98
94
95
THINKExcellent
7Apple M4openai/gpt-oss-20b20B/MXFP424 GBLM-STUDIOMLX29.31.8 s
99
93
95
THINKExcellent
8Cortex-X925gemma4:26b25.8B/Q4_K_M122 GBOLLAMAGGUF67.4529 ms
100
92
94
Excellent
9Intel Xeon® E5-2637 v3qwen3-30b-a3b-t16-c32k-instruct:latest30.5B/Q4_K_M96 GBOLLAMAGGUF116.0575 ms
100
91
94
Excellent
10Intel Core™ i7-14700FGPT-20B:20b20.9B/MXFP464 GBOLLAMAGGUF181.01.3 s
96
93
94
THINKExcellent
11Apple M4 Progpt-oss:20b20.9B/MXFP448 GBOLLAMAGGUF54.52.9 s
92
95
94
THINKExcellent
12Apple M4openai/gpt-oss-20b20B/MXFP432 GBLM-STUDIOMLX40.71.3 s
100
92
94
THINKExcellent
13Apple M4qwen/qwen3-30b-a3b-250730B/4bit32 GBLM-STUDIOMLX44.8405 ms
100
91
94
Excellent
14AMD RYZEN AI MAX+ 395gpt-oss:20b20.9B/MXFP464 GBOLLAMAGGUF48.02.3 s
91
95
94
THINKExcellent
15Cortex-X925gpt-oss:20b20.9B/MXFP4122 GBOLLAMAGGUF60.53.1 s
87
95
93
THINKExcellent
16Apple M4 Pronemotron-3-nano64 GBLM-STUDIOGGUF93.3257 ms
100
90
93
THINKExcellent
17AMD Ryzen 7 9700X 8-Core Processorgoogle/gemma-4-26b-a4b-qat26B/Q4_031 GBLM-STUDIOGGUF109.7156 ms
96
92
93
Excellent
18Intel Core™ i7-14700FGPT-OSS-20B:20b20.9B/MXFP4128 GBOLLAMAGGUF178.61.4 s
92
93
93
THINKExcellent
19AMD Ryzen 7 7800X3D 8-Core Processorgemma-4-26b-a4b-it-qat26B/Q4_K_XL31 GBLM-STUDIOGGUF166.9113 ms
93
93
93
ECOExcellent
20Intel Core™ i7-14700FGPT-20B:20b20.9B/MXFP4128 GBOLLAMAGGUF177.51.2 s
93
93
93
THINKExcellent
...

Don't miss new benchmarks

Get notified when new models and hardware configurations are tested. No spam, unsubscribe anytime.