Back to leaderboard

qwen3:4b

OLLAMA GGUF

qwen3 · 4.0B · Q4_K_M

Not Rec.

Nov 4, 2026 · Intel Core™ i7-9700

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.7B · Q4_K_M

Excellent

Mar 5, 2026 · Apple M1 Max

Global Score
29 vs 83
Hardware Fit
84 vs 100
Quality Score
6 vs 76

Hardware

qwen3:4b qwen3.5:4b
MachineVMware, Inc. VMware Virtual PlatformMacBook Pro
CPUIntel Core™ i7-9700Apple M1 Max
Cores410
RAM16 GB32 GB
GPUIntegrated / UnknownApple M1 Max
OSUbuntu 24.04.5 LTSmacOS 26.3
Archx64arm64
Power Modeunknownbalanced

Performance

qwen3:4b qwen3.5:4b
Tokens/sec10.432.4
First chunk857 ms306 ms
TTFT857 ms306 ms
Load time41.0 s2.7 s
Memory usage3.0 GB6.5 GB
Memory %19%20%

HW Fit Score Breakdown

qwen3:4b

Speed
34/50
TTFT
20/20
Memory
30/30

qwen3.5:4b

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

Quality

qwen3:4b

Reasoning
0/20
Coding
2/20
Instruction
2/20
Structured
0/15
Math
1/15
Multilingual
1/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor

qwen3.5:4b

Reasoning
16/20
Coding
14/20
Instruction
15/20
Structured
15/15
Math
7/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Weak Multilingual: Strong
View qwen3:4b details | View qwen3.5:4b 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