elephant-1 · IQ3_XS
This page aggregates 1 real-world runs of elephant-1 (IQ3_XS) on NVIDIA RTX PRO 6000 Blackwell 96GB with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
354.49 tok/s
Decode
Decode speed
1,856.14 tok/s
Prefill
Prefill speed
0.89 s
TTFT
Time to first token
13.82 GB
VRAM
VRAM usage
1
Measured runs
1
Independent sources
Other
Source platforms
13 days ago
Last verified
Performance
- Ollama · IQ1_S Decode 275.81 · Prefill 1195.7 ·
- Ollama · IQ4_XS Decode 194.74 · Prefill 889.56 ·
- Strata · IQ3_XS (current)Decode 354.49 · Prefill 1856.14 ·
- Strata · Q4_K_S Decode 293.11 · Prefill 1298.77 ·
Core figures
- Decode (avg)
- 354.49 tok/s
- Prefill (avg)
- 1,856.14 tok/s
- TTFT (avg)
- 0.89 s
- VRAM (avg)
- 13.82 GB
- MTP acceptance rate
- 0.6
- TTFB
- 2.84 GB
- Power draw
- 332.5 W
Configuration
Member-level fields are taken from the most recent run
- Model
- elephant-1
- Quantization
- IQ3_XS
- Framework
- Strata
- Version
- v0.26
- Context length
- 32768 tokens
- Batch size
- 512
- GPU layers
- 12
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA RTX PRO 6000 Blackwell 96GB
- Nominal VRAM
- 96 GB
- Measured VRAM (avg)
- 13.82 GB
- OS
- Windows 11 24H2
- Driver
- 560.94
- CUDA
- 12.4
- Power draw
- 332.5 W
Get started
Original model resources
Base model resources may not match this quantization. Use the verified deployment weights above when available.
Sources & evidence
1 measured records in total, each traceable to its original source
v0.26 · Windows 11 24H2 · CUDA 12.4 · 32768 ctx
354.49 tok/s
Decode
1,856.14 tok/s
Prefill
0.89 s
TTFT
13.82 GB
VRAM
0.6
MTP
332.5 W
Power draw
L2 batch seed 2026-10-22: Strata on hw_id=68 running model_id=95 at IQ3_XS