Back to leaderboard

qwen3:8b

OLLAMA GGUF

qwen3 · 8.2B · Q4_K_M

Excellent

Mar 5, 2026 · Intel Core™ Ultra 9 285K

qwen3.5-35b-a3b-t48-c8k:latest

OLLAMA GGUF

qwen35moe · 36.0B · Q4_K_M

Not Rec.

Sep 6, 2026 · Intel Xeon® E5-4650 v2

Global Score
84 vs 81
Hardware Fit
97 vs 63
Quality Score
79 vs 88

Hardware

qwen3:8b qwen3.5-35b-a3b-t4…
MachineASUSDell Inc. PowerEdge R820
CPUIntel Core™ Ultra 9 285KIntel Xeon® E5-4650 v2
Cores2480
RAM47 GB256 GB
GPUNVIDIA GeForce RTX 5090, Intel(R) GraphicsMicrosoft Remote Display Adapter, NVIDIA Tesla P4, NVIDIA Tesla P4, Microsoft Basic Display Adapter
OSMicrosoft Windows 11 Pro 10.0.26200Microsoft Windows Server 2022 Datacenter 10.0.20348
Archx64x64
Power Modebalancedbalanced

Performance

qwen3:8b qwen3.5-35b-a3b-t4…
Tokens/sec208.017.2
First chunk82 ms845 ms
TTFT1.3 s845 ms
Load time1.5 s103.1 s
Memory usage9.6 GB21.9 GB
Memory %20%9%

HW Fit Score Breakdown

qwen3:8b

Speed
50/50
TTFT
18/20
Memory
29/30

qwen3.5-35b-a3b-t48-c8k…

Speed
16/50
TTFT
17/20
Memory
30/30

Quality

qwen3:8b

Reasoning
18/20
Coding
10/20
Instruction
14/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong

qwen3.5-35b-a3b-t48-c8k…

Reasoning
17/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

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