Hy-Memory-30B-A3B · IQ2_XS
This page aggregates 1 real-world runs of Hy-Memory-30B-A3B (IQ2_XS) on NVIDIA RTX PRO 6000 Blackwell 96GB with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
349.83 tok/s
Decode
Decode speed
1,267.72 tok/s
Prefill
Prefill speed
1.3 s
TTFT
Time to first token
32.34 GB
VRAM
VRAM usage
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- llama.cpp · IQ2_M Decode 366.29 · Prefill 1896.32 ·
- llama.cpp · IQ3_XS Decode 355.22 · Prefill 1526.75 ·
- Strata · IQ2_XS (current)Decode 349.83 · Prefill 1267.72 ·
Core figures
- Decode (avg)
- 349.83 tok/s
- Prefill (avg)
- 1,267.72 tok/s
- TTFT (avg)
- 1.3 s
- VRAM (avg)
- 32.34 GB
- MTP acceptance rate
- 0.62
- TTFB
- 2.8 GB
- Power draw
- 362.7 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Hy-Memory-30B-A3B
- Quantization
- IQ2_XS
- Framework
- Strata
- Version
- v0.30
- Context length
- 32768 tokens
- Batch size
- 256
- GPU layers
- 8
- 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)
- 32.34 GB
- OS
- Windows 11 24H2
- Driver
- 565.57
- CUDA
- 12.6
- Power draw
- 362.7 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
v0.30 · Windows 11 24H2 · CUDA 12.6 · 32768 ctx
349.83 tok/s
Decode
1,267.72 tok/s
Prefill
1.3 s
TTFT
32.34 GB
VRAM
0.62
MTP
362.7 W
Power draw
L2 batch seed 2026-10-19: Strata on hw_id=68 running model_id=87 at IQ2_XS