Back to search results
DeepSeek-R1-Distill-Qwen-14B · IQ2_XS
NVIDIA H100 80GB HBM380GBOllama v0.40L2 Cross-framework verified
This page aggregates 1 real-world runs of DeepSeek-R1-Distill-Qwen-14B (IQ2_XS) on NVIDIA H100 80GB HBM3 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
291.58 tok/s
Decode
Decode speed
1,457.4 tok/s
Prefill
Prefill speed
2.21 s
TTFT
Time to first token
16.21 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
17 days ago
Last verified
Performance
- vLLM · IQ4_XS Decode 398.24 · Prefill 1414.76 ·
- vLLM · Q5_K_S Decode 364.79 · Prefill 1787.53 ·
- Ollama · IQ2_XS (current)Decode 291.58 · Prefill 1457.4 ·
- Ollama · IQ3_XS Decode 340.5 · Prefill 1221.18 ·
Core figures
- Decode (avg)
- 291.58 tok/s
- Prefill (avg)
- 1,457.4 tok/s
- TTFT (avg)
- 2.21 s
- VRAM (avg)
- 16.21 GB
- MTP acceptance rate
- —
- TTFB
- 2.33 GB
- Power draw
- 601.5 W
Configuration
Member-level fields are taken from the most recent run
- Model
- DeepSeek-R1-Distill-Qwen-14B
- Quantization
- IQ2_XS
- Framework
- Ollama
- Version
- v0.40
- Context length
- 4096 tokens
- Batch size
- 512
- GPU layers
- 20
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA H100 80GB HBM3
- Nominal VRAM
- 80 GB
- Measured VRAM (avg)
- 16.21 GB
- OS
- Ubuntu 24.04
- Driver
- 565.57
- CUDA
- 12.8
- Power draw
- 601.5 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.40 · Ubuntu 24.04 · CUDA 12.8 · 4096 ctx
291.58 tok/s
Decode
1,457.4 tok/s
Prefill
2.21 s
TTFT
16.21 GB
VRAM
—
MTP
601.5 W
Power draw
L2 batch seed 2026-10-19: Ollama on hw_id=55 running model_id=71 at IQ2_XS