Back to search results
Qwen3.5-4B · FP16
NVIDIA RTX 508016GBOllama v0.36L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-4B (FP16) on NVIDIA RTX 5080 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
104.67 tok/s
Decode
Decode speed
436.92 tok/s
Prefill
Prefill speed
1.15 s
TTFT
Time to first token
15.2 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- Ollama · FP16 (current)Decode 104.67 · Prefill 436.92 ·
- Ollama · Q4_K_S Decode 194.55 · Prefill 763.79 ·
- Strata · Q8_0 Decode 218.51 · Prefill 985.54 ·
- llama.cpp · IQ1_S Decode 350.61 · Prefill 1829.73 ·
- llama.cpp · IQ2_XS Decode 312.6 · Prefill 1217.96 ·
- Strata · IQ3_S Decode 356.3 · Prefill 1901.07 ·
- llama.cpp · IQ4_XS Decode 133.14 · Prefill 731.5 ·
- Strata · Q3_K_M Decode 231.73 · Prefill 785.76 ·
Core figures
- Decode (avg)
- 104.67 tok/s
- Prefill (avg)
- 436.92 tok/s
- TTFT (avg)
- 1.15 s
- VRAM (avg)
- 15.2 GB
- MTP acceptance rate
- —
- TTFB
- 0.84 GB
- Power draw
- 309.8 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-4B
- Quantization
- FP16
- Framework
- Ollama
- Version
- v0.36
- Context length
- 16384 tokens
- Batch size
- 256
- GPU layers
- 20
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX 5080
- Nominal VRAM
- 16 GB
- Measured VRAM (avg)
- 15.2 GB
- OS
- Windows 11 23H2
- Driver
- 565.57
- CUDA
- 12.4
- Power draw
- 309.8 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.36 · Windows 11 23H2 · CUDA 12.4 · 16384 ctx
104.67 tok/s
Decode
436.92 tok/s
Prefill
1.15 s
TTFT
15.2 GB
VRAM
—
MTP
309.8 W
Power draw
L2 batch seed 2026-10-19: Ollama on hw_id=14 running model_id=48 at FP16