nemotron-3.5-lightning:latest

nemotron_h_moe · 32.9B · Q4_K_M

THINKING MODEL

ASUSTeK COMPUTER INC. GX10 (Cortex-X925)

122 GB · Ubuntu 24.04.4 LTS

Tested on August 17, 2026

Top 18% Compare

Global Score

87 /100

Not Rec.

Hardware Fit

80/100

Quality

90/100

Get this model

🦙

Ollama

ollama pull nemotron-3.5-lightning:latest

View on Ollama Library

ollama.com/library/nemotron-3.5-lightning

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: ASUSTeK COMPUTER INC. GX10
CPU: Cortex-X925
Cores: 20 total (20 perf)
Frequency: 3.35 GHz
RAM: 122 GB
GPU: Device 2e12
OS: Ubuntu 24.04.4 LTS
Arch: arm64
Power Mode: performance

Performance

Tokens/sec Tokens generated per second — higher is better: 84.1
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±4.9
First chunk latency Delay before the first streamed chunk arrives from runtime: 371 ms
Time to first token How fast the model starts responding — lower is better: 30.0 s
Load time Time to load the model into memory before inference: 0.3 s
Memory usage RAM consumed during inference — percentage of total system memory: 24.9 GB (20%)
Total tokens Total number of tokens generated across all benchmark prompts: 1461
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~787

Score breakdown

Speed

50/50

Time to first token

0/20

Memory

30/30

Quality

Reasoning

19/20

Coding

17/20

Instruction following

17/20

Structured output

13/15

Math

14/15

Multilingual

10/10

Category levels

Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.32.11
Model format: GGUF
Hardware profile: HIGH-END
Result hash: 672a9c01c21bff5a101bef3af866fc228634fb11b04d3aaea30c9b390bf6327a

Interpretation

Hardware fit: 80/100. Overall suitability: NOT RECOMMENDED (Global 87/100). Category profile: Reasoning: Strong, Coding: Strong, Instruction Following: Strong, Structured Output: Strong, Math: Strong, Multilingual: Strong.

Disqualifiers

Time to first token too high: 30000ms (maximum: 11500ms for HIGH-END profile)

Bench Environment

Thermal: nominal CPU load: avg 5% (peak 6%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest