Back to search results
Qwen3-8B · IQ4_XS
NVIDIA RTX 409024GBllama.cpp b4213L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-8B (IQ4_XS) on NVIDIA RTX 4090 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
112 tok/s
Decode
Decode speed
1,323 tok/s
Prefill
Prefill speed
1.4 s
TTFT
Time to first token
3.2 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
- llama.cpp · IQ4_XS (current)Decode 112 · Prefill 1323 ·
- vLLM · IQ4_XS Decode 104 · Prefill 1176 ·
- vLLM · Q4_K_M Decode 96 · Prefill 1061 ·
- vLLM · Q5_K_M Decode 95 · Prefill 815 ·
- vLLM · Q8_0 Decode 62 · Prefill 723 ·
- Ollama · Q4_K_M Decode 113 · Prefill 922 ·
- Ollama · Q5_K_M Decode 95 · Prefill 1090 ·
- llama.cpp · Q8_0 Decode 91.4 · Prefill 685.5 ·
- llama.cpp · Q5_K_M Decode 133.3 · Prefill 999.8 ·
- llama.cpp · Q4_K_M Decode 160.5 · Prefill 1203.8 ·
Core figures
- Decode (avg)
- 112 tok/s
- Prefill (avg)
- 1,323 tok/s
- TTFT (avg)
- 1.4 s
- VRAM (avg)
- 3.2 GB
- MTP acceptance rate
- —
- TTFB
- — GB
- Power draw
- 458 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-8B
- Quantization
- IQ4_XS
- Framework
- llama.cpp
- Version
- b4213
- Context length
- 8192 tokens
- Batch size
- 512
- GPU layers
- 99
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 4090
- Nominal VRAM
- 24 GB
- Measured VRAM (avg)
- 3.2 GB
- OS
- Linux 6.8.0
- Driver
- 570.86.15
- CUDA
- 12.6
- Power draw
- 458 W
Sources & evidence
1 measured records in total, each traceable to its original source
b4213 · Linux 6.8.0 · CUDA 12.6 · 8192 ctx
112 tok/s
Decode
1,323 tok/s
Prefill
1.4 s
TTFT
3.2 GB
VRAM
—
MTP
458 W
Power draw
Auto-generated LLM-sourced record 2026-10-09. Model: Qwen3-8B, Hardware: RTX 4090, Framework: 1, Quant: IQ4_XS.