Back to leaderboard

phi3:3.8b

OLLAMA GGUF

phi3 · 3.8B · Q4_0

Good

Mar 2, 2026 · Apple M4

qwen/qwen3.6-35b-a3b

LM-STUDIO MLX

qwen3_5_moe · 35B · 4bit

Excellent

Jul 17, 2026 · Apple M5 Max

Global Score
64 vs 89
Hardware Fit
100 vs 97
Quality Score
48 vs 86

Hardware

phi3:3.8b qwen/qwen3.6-35b-a3b
MachineMacBook AirMacBook Pro
CPUApple M4Apple M5 Max
Cores1018
RAM32 GB64 GB
GPUApple M4Apple M5 Max
OSmacOS 26.3macOS 27.0
Archarm64arm64
Power Modebalancedbalanced

Performance

phi3:3.8b qwen/qwen3.6-35b-a3b
Tokens/sec42.0130.2
First chunkN/A5 ms
TTFT142 ms135 ms
Load time1.1 sN/A
Memory usage16.0 GB26.6 GB
Memory %50%42%

HW Fit Score Breakdown

phi3:3.8b

Speed
40/50
TTFT
30/20
Memory
30/30

qwen/qwen3.6-35b-a3b

Speed
50/50
TTFT
20/20
Memory
27/30

Quality

phi3:3.8b

Reasoning
12/20
Coding
3/20
Instruction
11/20
Structured
10/15
Math
5/15
Multilingual
7/10
Reasoning: Adequate Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Weak Multilingual: Adequate

qwen/qwen3.6-35b-a3b

Reasoning
15/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong 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