Back to leaderboard

deepseek-r1:latest

OLLAMA GGUF

qwen2 · 7.6B · Q4_K_M

Not Rec.

Jun 26, 2026 · Cortex-A78AE

qwen3.5-9b-mlx

LM-STUDIO MLX

qwen3_5 · 9B · 4bit

Marginal

Mar 11, 2026 · Apple M4

Global Score
22 vs 59
Hardware Fit
69 vs 65
Quality Score
2 vs 57

Hardware

deepseek-r1:latest qwen3.5-9b-mlx
MachineNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit SuperMac mini
CPUCortex-A78AEApple M4
Cores610
RAM7 GB24 GB
GPUIntegrated / UnknownApple M4
OSUbuntu 22.04.5 LTSmacOS 26.3
Archarm64arm64
Power Modebalancedbalanced

Performance

deepseek-r1:latest qwen3.5-9b-mlx
Tokens/sec9.614.3
First chunk1300 ms24 ms
TTFT1.3 s13.4 s
Load time19.9 s13.3 s
Memory usage4.8 GB2.5 GB
Memory %65%10%

HW Fit Score Breakdown

deepseek-r1:latest

Speed
31/50
TTFT
20/20
Memory
18/30

qwen3.5-9b-mlx

Speed
32/50
TTFT
3/20
Memory
30/30

Quality

deepseek-r1:latest

Reasoning
0/20
Coding
1/20
Instruction
0/20
Structured
0/15
Math
1/15
Multilingual
0/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor

qwen3.5-9b-mlx

Reasoning
10/20
Coding
17/20
Instruction
5/20
Structured
9/15
Math
11/15
Multilingual
5/10
Reasoning: Weak Coding: Strong Instruction Following: Weak Structured Output: Adequate Math: Adequate Multilingual: Adequate
View deepseek-r1:latest details | View qwen3.5-9b-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