qwen3.5-0.8b-mlx
LM-STUDIO MLXqwen3_5 · 0.8B · 8bit
Mar 3, 2026 · Apple M4
falcon-h1-0.5b-instruct
LM-STUDIO GGUFfalcon-h1 · 0.5B · Q4_K_M
Mar 8, 2026 · Intel Core™ i5-5300U
Global Score
60 vs 57
Hardware Fit
96 vs 91
Quality Score
44 vs 42
Hardware
qwen3.5-0.8b-mlx falcon-h1-0.5b-ins…
MachineMacBook AirLENOVO 20BUS00700
CPUApple M4Intel Core™ i5-5300U
Cores104
RAM32 GB16 GB
GPUApple M4Intel(R) HD Graphics 5500
OSmacOS 26.3Microsoft Windows 10 Professionnel 10.0.19045
Archarm64x64
Power Modebalancedbalanced
Performance
qwen3.5-0.8b-mlx falcon-h1-0.5b-ins…
Tokens/sec108.016.3
First chunk2568 ms2987 ms
TTFT2.6 s3.0 s
Load timeN/AN/A
Memory usage1.0 GB0.3 GB
Memory %3%2%
HW Fit Score Breakdown
qwen3.5-0.8b-mlx
Speed
50/50
TTFT
16/20
Memory
30/30
falcon-h1-0.5b-instruct
Speed
48/50
TTFT
16/20
Memory
27/30
Quality
qwen3.5-0.8b-mlx
Reasoning
10/20
Coding
3/20
Instruction
11/20
Structured
12/15
Math
3/15
Multilingual
5/10
Reasoning: Weak Coding: Poor Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Adequate
falcon-h1-0.5b-instruct
Reasoning
8/20
Coding
7/20
Instruction
12/20
Structured
10/15
Math
4/15
Multilingual
1/10
Reasoning: Weak Coding: Weak Instruction Following: Adequate Structured Output: Adequate Math: Weak Multilingual: Poor
Run yours and compare
$
npm install -g metrillm@latest$
metrillmRequires Node 20+ and Ollama or LM Studio running
Or run without installing: npx metrillm@latest