Back to leaderboard

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.7B · Q4_K_M

Good

Mar 5, 2026 · Intel Core™ i5-6500

codestral:22b

OLLAMA GGUF

llama · 22.2B · Q4_0

Good

Jun 26, 2026 · Cortex-X925

Global Score
61 vs 67
Hardware Fit
69 vs 67
Quality Score
58 vs 67

Hardware

qwen3.5:4b codestral:22b
MachineGigabyte Technology Co., Ltd. H170M-D3HNVIDIA NVIDIA_DGX_Spark
CPUIntel Core™ i5-6500Cortex-X925
Cores420
RAM16 GB122 GB
GPUGP104 [GeForce GTX 1070]Device 2e12
OSUbuntu 24.04.4 LTSUbuntu 24.04.4 LTS
Archx64arm64
Power Modelow-powerperformance

Performance

qwen3.5:4b codestral:22b
Tokens/sec35.717.2
First chunk438 ms129 ms
TTFT5.1 s129 ms
Load time40.1 s11.6 s
Memory usage5.5 GB18.9 GB
Memory %35%16%

HW Fit Score Breakdown

qwen3.5:4b

Speed
50/50
TTFT
11/20
Memory
8/30

codestral:22b

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

Quality

qwen3.5:4b

Reasoning
18/20
Coding
1/20
Instruction
8/20
Structured
8/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Poor Instruction Following: Weak Structured Output: Adequate Math: Strong Multilingual: Strong

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