Back to leaderboard

mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Oct 20, 2026 · Apple M1 Pro

qwen/qwen3.5-35b-a3b

LM-STUDIO GGUF

qwen35moe · 35B · Q6_K

Not Rec.

Sep 19, 2026 · AMD Ryzen 7 5800X 8-Core Processor

Global Score
73 vs 72
Hardware Fit
64 vs 35
Quality Score
77 vs 88

Hardware

mistralai/devstral… qwen/qwen3.5-35b-a3b
MachineMacBook ProSystem manufacturer
CPUApple M1 ProAMD Ryzen 7 5800X 8-Core Processor
Cores1016
RAM32 GB32 GB
GPUApple M1 ProAMD Radeon RX 9060 XT
OSmacOS 15.6.1Microsoft Windows 11 Home 10.0.26200
Archarm64x64
Power Modebalancedbalanced

Performance

mistralai/devstral… qwen/qwen3.5-35b-a3b
Tokens/sec11.98.3
First chunk7 ms32 ms
TTFT1.1 s2.0 s
Load timeN/AN/A
Memory usage19.7 GB28.9 GB
Memory %62%91%

HW Fit Score Breakdown

mistralai/devstral-smal…

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

qwen/qwen3.5-35b-a3b

Speed
13/50
TTFT
17/20
Memory
5/30

Quality

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

qwen/qwen3.5-35b-a3b

Reasoning
16/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View mistralai/devstral… details | View qwen/qwen3.5-35b-a3b 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