Back to search results
Qwen3.5-4B · Q4_K_M
NVIDIA RTX 20606GBllama.cpp b11206L1 Reproduced
This page aggregates 1 real-world runs of Qwen3.5-4B (Q4_K_M) on NVIDIA RTX 2060 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
56.1 tok/s
Decode
Decode speed
131 tok/s
Prefill
Prefill speed
— s
TTFT
Time to first token
— GB
VRAM
VRAM usage
L1 Reproduced
1
Measured runs
1
Independent sources
GitHub
Source platforms
Yesterday
Last verified
Performance
- llama.cpp · Q4_K_M (current)Decode 56.1 · Prefill 131 ·
Core figures
- Decode (avg)
- 56.1 tok/s
- Prefill (avg)
- 131 tok/s
- TTFT (avg)
- — s
- VRAM (avg)
- — GB
- MTP acceptance rate
- —
- TTFB
- — GB
- Power draw
- — W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-4B
- Quantization
- Q4_K_M
- Framework
- llama.cpp
- Version
- b11206
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 2060
- Nominal VRAM
- 6 GB
- OS
- Windows 11
Sources & evidence
1 measured records in total, each traceable to its original source
b11206 · Windows 11
56.1 tok/s
Decode
131 tok/s
Prefill
— s
TTFT
0 GB
VRAM
—
MTP
— W
Power draw
CUDA | 3rd+ | 117-131 tok/s | 56.1 tok/s