Back to leaderboard

qwen3.5-0.8b-mlx

LM-STUDIO MLX

qwen3_5 · 0.8B · 4bit

Marginal

Mar 6, 2026 · Apple M4 Pro

gemma-4-26b-a4b-it-qat

LM-STUDIO GGUF

gemma4 · 26B · Q4_K_XL

Excellent

Jul 7, 2026 · AMD Ryzen 7 7800X3D 8-Core Processor

Global Score
53 vs 90
Hardware Fit
89 vs 82
Quality Score
38 vs 93

Hardware

qwen3.5-0.8b-mlx gemma-4-26b-a4b-it…
MachineMac miniASUS
CPUApple M4 ProAMD Ryzen 7 7800X3D 8-Core Processor
Cores1416
RAM64 GB31 GB
GPUApple M4 ProNVIDIA GeForce RTX 4090, Raphael
OSmacOS 15.7.4cachyos rolling
Archarm64x64
Power Modebalancedlow-power

Performance

qwen3.5-0.8b-mlx gemma-4-26b-a4b-it…
Tokens/sec52.9156.5
First chunk4013 ms10 ms
TTFT4.0 s124 ms
Load timeN/AN/A
Memory usage0.6 GB23.3 GB
Memory %1%76%

HW Fit Score Breakdown

qwen3.5-0.8b-mlx

Speed
50/50
TTFT
9/20
Memory
30/30

gemma-4-26b-a4b-it-qat

Speed
50/50
TTFT
20/20
Memory
12/30

Quality

qwen3.5-0.8b-mlx

Reasoning
7/20
Coding
2/20
Instruction
12/20
Structured
10/15
Math
2/15
Multilingual
5/10
Reasoning: Weak Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Poor Multilingual: Adequate

gemma-4-26b-a4b-it-qat

Reasoning
19/20
Coding
19/20
Instruction
18/20
Structured
14/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View qwen3.5-0.8b-mlx details | View gemma-4-26b-a4b-it… 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