Back to search results
Qwen3-1.7B · IQ2_M
NVIDIA Jetson AGX Orin 64GB64GBStrata v0.50L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-1.7B (IQ2_M) on NVIDIA Jetson AGX Orin 64GB with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
189.25 tok/s
Decode
Decode speed
901.13 tok/s
Prefill
Prefill speed
0.84 s
TTFT
Time to first token
3.96 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- vLLM · FP16 Decode 57.65 · Prefill 238.63 ·
- vLLM · IQ3_XS Decode 131.36 · Prefill 565.49 ·
- Strata · IQ2_M (current)Decode 189.25 · Prefill 901.13 ·
- Strata · Q3_K_M Decode 175.11 · Prefill 755.7 ·
- Ollama · IQ2_M Decode 124.51 · Prefill 592.75 ·
- Ollama · IQ3_S Decode 119.35 · Prefill 527.63 ·
- Strata · IQ4_XS Decode 159.91 · Prefill 540.21 ·
Core figures
- Decode (avg)
- 189.25 tok/s
- Prefill (avg)
- 901.13 tok/s
- TTFT (avg)
- 0.84 s
- VRAM (avg)
- 3.96 GB
- MTP acceptance rate
- 0.7
- TTFB
- 1.51 GB
- Power draw
- 29.3 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-1.7B
- Quantization
- IQ2_M
- Framework
- Strata
- Version
- v0.50
- Context length
- 8192 tokens
- Batch size
- 1024
- GPU layers
- 5
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA Jetson AGX Orin 64GB
- Nominal VRAM
- 64 GB
- Measured VRAM (avg)
- 3.96 GB
- OS
- JetPack 6.0
- Driver
- NVIDIA L4T 36.3
- CUDA
- 12.6
- Power draw
- 29.3 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.50 · JetPack 6.0 · CUDA 12.6 · 8192 ctx
189.25 tok/s
Decode
901.13 tok/s
Prefill
0.84 s
TTFT
3.96 GB
VRAM
0.7
MTP
29.3 W
Power draw
L2 batch seed 2026-10-16: Strata on hw_id=76 running model_id=55 at IQ2_M