Back to leaderboard

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.2B · Q4_K_M

Marginal

Oct 12, 2026 · AMD Ryzen 5 5500

codestral:22b

OLLAMA GGUF

llama · 22.2B · Q4_0

Good

Jun 26, 2026 · Cortex-X925

Global Score
50 vs 67
Hardware Fit
93 vs 67
Quality Score
31 vs 67

Hardware

qwen3.5:4b codestral:22b
MachineMicrosoft WSLNVIDIA NVIDIA_DGX_Spark
CPUAMD Ryzen 5 5500Cortex-X925
Cores1220
RAM8 GB122 GB
GPUIntegrated / UnknownDevice 2e12
OSUbuntu 26.04.1 LTSUbuntu 24.04.4 LTS
Archx64arm64
Power Modeunknownperformance

Performance

qwen3.5:4b codestral:22b
Tokens/sec92.617.2
First chunk323 ms129 ms
TTFT2.8 s129 ms
Load time9.0 s11.6 s
Memory usage2.9 GB18.9 GB
Memory %38%16%

HW Fit Score Breakdown

qwen3.5:4b

Speed
50/50
TTFT
15/20
Memory
28/30

codestral:22b

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

Quality

qwen3.5:4b

Reasoning
18/20
Coding
1/20
Instruction
0/20
Structured
0/15
Math
12/15
Multilingual
0/10
Reasoning: Strong Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Strong Multilingual: Poor

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
View qwen3.5:4b details | View codestral:22b 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