Back to search results
Qwen3.5-0.8B · IQ3_XS
NVIDIA RTX 4070 Ti12GBllama.cpp v0.29L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-0.8B (IQ3_XS) on NVIDIA RTX 4070 Ti with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
242.17 tok/s
Decode
Decode speed
959.61 tok/s
Prefill
Prefill speed
2.02 s
TTFT
Time to first token
3.14 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
5 days ago
Last verified
Performance
- llama.cpp · IQ4_XS Decode 272.81 · Prefill 1057.01 ·
- vLLM · IQ3_S Decode 341.56 · Prefill 1371.63 ·
- vLLM · Q8_0 Decode 188.8 · Prefill 665.28 ·
- llama.cpp · IQ3_XS (current)Decode 242.17 · Prefill 959.61 ·
- llama.cpp · Q3_K_S Decode 230.59 · Prefill 747.57 ·
Core figures
- Decode (avg)
- 242.17 tok/s
- Prefill (avg)
- 959.61 tok/s
- TTFT (avg)
- 2.02 s
- VRAM (avg)
- 3.14 GB
- MTP acceptance rate
- —
- TTFB
- 1.94 GB
- Power draw
- 199.1 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-0.8B
- Quantization
- IQ3_XS
- Framework
- llama.cpp
- Version
- v0.29
- Context length
- 8192 tokens
- Batch size
- 512
- GPU layers
- 28
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 4070 Ti
- Nominal VRAM
- 12 GB
- Measured VRAM (avg)
- 3.14 GB
- OS
- Ubuntu 22.04
- Driver
- 570.86
- CUDA
- 12.8
- Power draw
- 199.1 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.29 · Ubuntu 22.04 · CUDA 12.8 · 8192 ctx
242.17 tok/s
Decode
959.61 tok/s
Prefill
2.02 s
TTFT
3.14 GB
VRAM
—
MTP
199.1 W
Power draw
L2 batch seed 2026-10-13: llama.cpp on hw_id=42 running model_id=51 at IQ3_XS