Back to leaderboard

mistral:latest

OLLAMA GGUF

llama · 7.2B · Q4_K_M

Good

May 27, 2026 · AMD Ryzen 5 2600X Six-Core Processor

qwen3.5-4b-mlx

LM-STUDIO MLX

qwen3_5 · 4B · 4bit

Not Rec.

Mar 6, 2026 · Apple M4 Pro

Global Score
73 vs 62
Hardware Fit
100 vs 35
Quality Score
62 vs 73

Hardware

mistral:latest qwen3.5-4b-mlx
MachineMicro-Star International Co., Ltd. MS-7B79Mac mini
CPUAMD Ryzen 5 2600X Six-Core ProcessorApple M4 Pro
Cores1214
RAM16 GB64 GB
GPUNVIDIA GeForce GTX 1080Apple M4 Pro
OSMicrosoft Windows 11 Famille 10.0.26200macOS 15.7.4
Archx64arm64
Power Modebalancedbalanced

Performance

mistral:latest qwen3.5-4b-mlx
Tokens/sec31.70.0
First chunk102 ms5548 ms
TTFT102 ms5.5 s
Load time14.3 sN/A
Memory usage4.8 GB2.9 GB
Memory %30%5%

HW Fit Score Breakdown

mistral:latest

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

qwen3.5-4b-mlx

Speed
0/50
TTFT
6/20
Memory
29/30

Quality

mistral:latest

Reasoning
8/20
Coding
13/20
Instruction
16/20
Structured
12/15
Math
5/15
Multilingual
8/10
Reasoning: Weak Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Weak Multilingual: Strong

qwen3.5-4b-mlx

Reasoning
17/20
Coding
17/20
Instruction
9/20
Structured
8/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Adequate Math: Strong Multilingual: Strong

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