Back to leaderboard

qwen3.8:27b

OLLAMA GGUF

qwen35 · 27.3B · Q4_K_M

Excellent

Oct 3, 2026 · Cortex-X925

mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Oct 20, 2026 · Apple M1 Pro

Global Score
84 vs 73
Hardware Fit
66 vs 64
Quality Score
92 vs 77

Hardware

qwen3.8:27b mistralai/devstral…
MachineNVIDIA NVIDIA_DGX_SparkMacBook Pro
CPUCortex-X925Apple M1 Pro
Cores2010
RAM122 GB32 GB
GPUDevice 2e12Apple M1 Pro
OSUbuntu 24.04.5 LTSmacOS 15.6.1
Archarm64arm64
Power Modeperformancebalanced

Performance

qwen3.8:27b mistralai/devstral…
Tokens/sec29.411.9
First chunk306 ms7 ms
TTFT4.8 s1.1 s
Load time5.3 sN/A
Memory usage16.2 GB19.7 GB
Memory %13%62%

HW Fit Score Breakdown

qwen3.8:27b

Speed
31/50
TTFT
5/20
Memory
30/30

mistralai/devstral-smal…

Speed
25/50
TTFT
20/20
Memory
19/30

Quality

qwen3.8:27b

Reasoning
19/20
Coding
19/20
Instruction
14/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong

mistralai/devstral-smal…

Reasoning
17/20
Coding
16/20
Instruction
13/20
Structured
14/15
Math
7/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Weak 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