Back to leaderboard

qwen2.5-coder:3b

OLLAMA GGUF

qwen2 · 3.1B · Q4_K_M

Good

Jun 26, 2026 · Cortex-A78AE

qwen3.5:4b-mlx

OLLAMA GGUF

nvfp4

Good

Aug 25, 2026 · Apple M1 Pro

Global Score
73 vs 68
Hardware Fit
100 vs 91
Quality Score
61 vs 58

Hardware

qwen2.5-coder:3b qwen3.5:4b-mlx
MachineNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit SuperMacBook Pro
CPUCortex-A78AEApple M1 Pro
Cores610
RAM7 GB16 GB
GPUIntegrated / UnknownApple M1 Pro
OSUbuntu 22.04.5 LTSmacOS 26.6.2
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen2.5-coder:3b qwen3.5:4b-mlx
Tokens/sec21.342.2
First chunk969 ms201 ms
TTFT969 ms4.4 s
Load time6.6 s0.0 s
Memory usage2.0 GB3.8 GB
Memory %27%24%

HW Fit Score Breakdown

qwen2.5-coder:3b

Speed
50/50
TTFT
20/20
Memory
30/30

qwen3.5:4b-mlx

Speed
50/50
TTFT
11/20
Memory
30/30

Quality

qwen2.5-coder:3b

Reasoning
8/20
Coding
15/20
Instruction
11/20
Structured
14/15
Math
4/15
Multilingual
9/10
Reasoning: Weak Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong

qwen3.5:4b-mlx

Reasoning
16/20
Coding
2/20
Instruction
13/20
Structured
7/15
Math
11/15
Multilingual
9/10
Reasoning: Strong Coding: Poor Instruction Following: Adequate Structured Output: Weak Math: Adequate Multilingual: Strong
View qwen2.5-coder:3b details | View qwen3.5:4b-mlx details

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest