Back to leaderboard

llama-3.2-1b-instruct

LM-STUDIO GGUF

llama · 1B · Q4_K_M

Marginal

Mar 5, 2026 · Intel Core™ i5-5300U

qwen3.5-4b-mlx

LM-STUDIO MLX

qwen3_5 · 4B · 4bit

Good

Mar 3, 2026 · Apple M4

Global Score
50 vs 71
Hardware Fit
79 vs 88
Quality Score
37 vs 63

Hardware

llama-3.2-1b-instr… qwen3.5-4b-mlx
MachineLENOVO 20BUS00700MacBook Air
CPUIntel Core™ i5-5300UApple M4
Cores410
RAM16 GB32 GB
GPUIntel(R) HD Graphics 5500Apple M4
OSMicrosoft Windows 10 Professionnel 10.0.19045macOS 26.3
Archx64arm64
Power Modebalancedbalanced

Performance

llama-3.2-1b-instr… qwen3.5-4b-mlx
Tokens/sec9.923.1
First chunk1668 ms3279 ms
TTFT1.7 s3.3 s
Load timeN/AN/A
Memory usage0.8 GB2.9 GB
Memory %5%9%

HW Fit Score Breakdown

llama-3.2-1b-instruct

Speed
32/50
TTFT
20/20
Memory
27/30

qwen3.5-4b-mlx

Speed
45/50
TTFT
13/20
Memory
30/30

Quality

llama-3.2-1b-instruct

Reasoning
4/20
Coding
3/20
Instruction
10/20
Structured
10/15
Math
2/15
Multilingual
8/10
Reasoning: Poor Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Poor Multilingual: Strong

qwen3.5-4b-mlx

Reasoning
15/20
Coding
16/20
Instruction
6/20
Structured
6/15
Math
11/15
Multilingual
9/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak Structured Output: Weak Math: Adequate Multilingual: Strong
View llama-3.2-1b-instr… details | View qwen3.5-4b-mlx 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