§ 001 / MEASURED FIT RECEIPT

RTX 3070 8GB runs Qwen3.5 9B.

One recorded result from owned hardware. The number below is tied to the quant, context, workload, and measurement date shown.

Try this on my rig Preview receipt card Open receipt markdown

GPU

RTX 3070 8GB

Model tag

qwen3.5:9b

Generation

59.39 tok/s

Measured

2026-07-26

§ 002 / MEASURED ON REAL HARDWARE

Headline run

Q4_K_M · 4K context

Workload

Agent code task · 512 tokens

Prompt eval

18.70 tok/s

Peak VRAM

6.8 GB

Source date

2026-07-26

Run command

ollama run qwen3.5:9b

§ 003 / ALL RECORDED RUNS

All recorded runs

Context changes the memory budget and speed. A result measured at 4K is not a measured claim at 8K.

Measured runs for RTX 3070 8GB and Qwen3.5 9B
Context	Workload	Quant	Generation	Prompt	Peak VRAM
4K context	Agent code task · 512 tokens	Q4_K_M	59.39 tok/s	18.70	6.8 GB

§ 004 / CITE THIS RUN

Cite this measured run

BMD Pat measured RTX 3070 8GB running qwen3.5:9b (Q4_K_M, 4096 context tokens, Agent code task · 512 tokens): 59.39 generation tok/s and 6.8 GB peak VRAM on 2026-07-26. Source: https://bmdpat.com/receipts/rtx-3070/qwen-3-5-9b

Runtime version and model revision are not recorded in this catalog row. Keep that limit with the citation. Speed and memory measurements do not establish answer quality.

§ 005 / EVIDENCE AND NEXT STEP

Evidence and next step

This receipt is a measured data point, not a promise that every runtime or driver behaves identically. The source record is kept below so the claim can be audited and extended with the next run.

Reports/Sizing-Desk/fluarmn-tier-8gb-2026-07-25.jsonl

Enter your GPU and save your rig for future model drops →