qwen3.5-35b-a3b-t48-c8k:latest

OLLAMA GGUF

qwen35moe · 36.0B · Q4_K_M

Not Rec.

Jul 30, 2026 · Intel Xeon® E5-4620 0

mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Oct 20, 2026 · Apple M1 Pro

Global Score

10 vs 73

Hardware Fit

33 vs 64

Quality Score

0 vs 77

Hardware

qwen3.5-35b-a3b-t4… mistralai/devstral…

MachineDell Inc. PowerEdge R820MacBook Pro

CPUIntel Xeon® E5-4620 0Apple M1 Pro

Cores6410

RAM256 GB32 GB

GPUMicrosoft Remote Display Adapter, Microsoft Basic Display AdapterApple M1 Pro

OSMicrosoft Windows Server 2022 Datacenter 10.0.20348macOS 15.6.1

Archx64arm64

Power Modebalancedbalanced

Performance

qwen3.5-35b-a3b-t4… mistralai/devstral…

Tokens/sec3.511.9

First chunk3509 ms7 ms

TTFT30.0 s1.1 s

Load time30.0 sN/A

Memory usage22.0 GB19.7 GB

Memory %9%62%

HW Fit Score Breakdown

qwen3.5-35b-a3b-t48-c8k…

Speed

3/50

TTFT

0/20

Memory

30/30

mistralai/devstral-smal…

Speed

25/50

TTFT

20/20

Memory

19/30

Quality

qwen3.5-35b-a3b-t48-c8k…

Reasoning

0/20

Coding

0/20

Instruction

0/20

Structured

0/15

Math

0/15

Multilingual

0/10

Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor

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

View qwen3.5-35b-a3b-t4… details | View mistralai/devstral… 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