Back to search results
Llama-3-8B-Instruct · Q3_K_L
AMD Ryzen AI Max+ 395128GBOllama v0.38L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3-8B-Instruct (Q3_K_L) on AMD Ryzen AI Max+ 395 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
37.4 tok/s
Decode
Decode speed
165.99 tok/s
Prefill
Prefill speed
0.97 s
TTFT
Time to first token
14.2 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- vLLM · IQ2_M Decode 48.9 · Prefill 230.21 ·
- vLLM · Q5_K_M Decode 31.8 · Prefill 157.2 ·
- Ollama · Q3_K_L (current)Decode 37.4 · Prefill 165.99 ·
Core figures
- Decode (avg)
- 37.4 tok/s
- Prefill (avg)
- 165.99 tok/s
- TTFT (avg)
- 0.97 s
- VRAM (avg)
- 14.2 GB
- MTP acceptance rate
- —
- TTFB
- 0.3 GB
- Power draw
- 62.9 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3-8B-Instruct
- Quantization
- Q3_K_L
- Framework
- Ollama
- Version
- v0.38
- Context length
- 2048 tokens
- Batch size
- 512
- GPU layers
- 28
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- AMD Ryzen AI Max+ 395
- Nominal VRAM
- 128 GB
- Measured VRAM (avg)
- 14.2 GB
- OS
- Ubuntu 22.04
- Driver
- amdgpu 6.11
- CUDA
- 12.4
- Power draw
- 62.9 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.38 · Ubuntu 22.04 · CUDA 12.4 · 2048 ctx
37.4 tok/s
Decode
165.99 tok/s
Prefill
0.97 s
TTFT
14.2 GB
VRAM
—
MTP
62.9 W
Power draw
L2 batch seed 2026-10-19: Ollama on hw_id=24 running model_id=69 at Q3_K_L