Back to leaderboard

hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ2_XXS

OLLAMA GGUF

qwen35 · 27.3B · unknown

Not Rec.

Aug 22, 2026 · Apple M1 Pro

codestral:22b

OLLAMA GGUF

llama · 22.2B · Q4_0

Good

Jun 26, 2026 · Cortex-X925

Global Score
31 vs 67
Hardware Fit
37 vs 67
Quality Score
28 vs 67

Hardware

hf.co/unsloth/Qwen… codestral:22b
MachineMacBook ProNVIDIA NVIDIA_DGX_Spark
CPUApple M1 ProCortex-X925
Cores1020
RAM16 GB122 GB
GPUApple M1 ProDevice 2e12
OSmacOS 26.6.2Ubuntu 24.04.4 LTS
Archarm64arm64
Power Modebalancedperformance

Performance

hf.co/unsloth/Qwen… codestral:22b
Tokens/sec5.417.2
First chunk5076 ms129 ms
TTFT5.1 s129 ms
Load time0.3 s11.6 s
Memory usage11.6 GB18.9 GB
Memory %72%16%

HW Fit Score Breakdown

hf.co/unsloth/Qwen3.8-2…

Speed
13/50
TTFT
10/20
Memory
14/30

codestral:22b

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

Quality

hf.co/unsloth/Qwen3.8-2…

Reasoning
9/20
Coding
2/20
Instruction
1/20
Structured
1/15
Math
9/15
Multilingual
6/10
Reasoning: Weak Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Adequate Multilingual: Adequate

codestral:22b

Reasoning
11/20
Coding
17/20
Instruction
9/20
Structured
14/15
Math
6/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak 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