Back to leaderboard

llama3.3:70b

OLLAMA GGUF

llama · 70.6B · Q4_K_M

Not Rec.

Jun 26, 2026 · Cortex-X925

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
51 vs 35
Quality Score
83 vs 73

Hardware

llama3.3:70b qwen3.5-4b-mlx
MachineNVIDIA NVIDIA_DGX_SparkMac mini
CPUCortex-X925Apple M4 Pro
Cores2014
RAM122 GB64 GB
GPUDevice 2e12Apple M4 Pro
OSUbuntu 24.04.4 LTSmacOS 15.7.4
Archarm64arm64
Power Modeperformancebalanced

Performance

llama3.3:70b qwen3.5-4b-mlx
Tokens/sec4.60.0
First chunk560 ms5548 ms
TTFT560 ms5.5 s
Load time13.8 sN/A
Memory usage49.5 GB2.9 GB
Memory %41%5%

HW Fit Score Breakdown

llama3.3:70b

Speed
4/50
TTFT
20/20
Memory
27/30

qwen3.5-4b-mlx

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

Quality

llama3.3:70b

Reasoning
17/20
Coding
18/20
Instruction
12/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Adequate 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