#2772 Community benchmark: RTX 5070 Ti, i5-13600KF, 64 GB RAM, IQ3_S @ 262K context

本文

# Community benchmark on NVIDIA GeForce RTX 5070 Ti

Measured on 2026-10-16 with `tools/strata_bench.py` (suite 1, 3 runs per test, medians with ranges). Speeds are the engine's own timings; first-token times are measured by the client over HTTP. Speed only: no correctness check was run.

## Hardware and software

- GPU and VRAM: NVIDIA GeForce RTX 5070 Ti, 16 GiB VRAM
- CPU: 13th Gen Intel(R) Core(TM) i5-13600KF (14 cores, 20 threads); installed RAM: 63.8 GiB; storage: not measured
- OS: Windows 10 (10.0.19045); GPU driver: not read; CUDA: not read
- Strata: project 0.1.43.1, commit not a git checkout; engine 0.1.43.1
- Background workloads and power limits: not recorded

## Model and configuration

- Model and quantization: qwen3.8-flash-next-iq3_s
- Context: 262144 tokens; parallel slots: 1
- Settings: the install's own configuration, unchanged; thinking off for the suite; temperature 0

## Method

Run by `python tools/strata_bench.py` against the HTTP API of a Strata server (already running); model loading is not timed. One warm-up chat and a 1K prompt are discarded first. Every prompt starts with a unique tag, so the conversation cache cannot help the fresh-prompt tests.

- short chat: the three questions of `tools/ab_engine.py`, up to 100 new tokens, 3 runs
- 4K / 16K prompts: generated word lists sized by a calibration request, 24 new tokens, 3 fresh prompts each
- agent session: system prompt and four tools, then 5 turns that each add about 1500 words of a project file and continue from the model's own earlier answers (the way agent clients resend a conversation), 64 new tokens per turn, 3 sessions; the cache numbers come from the answer's usage (`cached_tokens`) and timings (`prompt_n`)
- concurrency (when run): N requests sent at the same moment, total tokens over the wall time of the burst, latency over all requests

## Results

| Configuration | Actual prompt tokens | Reused tokens | Generated tokens | Runs | Prompt tok/s median and range | Decode tok/s median and range | TTFT seconds median and range |
| --- | ---: | ---: | ---: | ---: | --- | --- | --- |
| short chat | 46 (43-51) | 0 | 100 (100-100) | 9 | not measured | 38.1 (32.6-42.9) | 0.95 (0.83-1.04) |
| 4K prompt, fresh | 4019 (4019-4019) | 0 (0-0) | 24 cap | 3 | 931.9 (857.1-947.8) | not measured | 4.36 (4.28-4.73) |
| 16K prompt, fresh | 16058 (16058-16058) | 0 (0-0) | 24 cap | 3 | 1199.6 (1198.5-1204.8) | not measured | 13.44 (13.38-13.47) |
| agent session, follow-up turns | 14487 (14440-14504) at the last turn | see below | up to 64 | 3 | see below | 42.2 (34.8-54.9) | 3.94 (3.60-4.14) |

Ranges are the lowest and highest of the runs; `Runs` counts requests that returned a token.

### Agent session (conversation cache)

Cache hit over the follow-up turns: 68.5 % (68.5-68.8); earlier-turn tokens read again per session: 144 (4-144) (earlier turns that the cache did not hold; the 13634 (13634-13635) new tokens of the files are read once); first turn (nothing cached) first token: 2.51 s (2.49-2.52).

| Turn (run 1) | Prompt tokens | Reused | Read | Prompt tok/s | First token (s) | Decode tok/s |
| ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| 0 | 609 | 0 | 609 | 249.8 | 2.486 | 31.9 |
| 1 | 3713 | 630 | 3083 | 773.9 | 4.042 | 37.3 |
| 2 | 5246 | 3776 | 1470 | 413.8 | 3.605 | 34.8 |
| 3 | 8690 | 5310 | 3380 | 827.8 | 4.143 | 38.5 |
| 4 | 11576 | 8716 | 2860 | 731.4 | 3.973 | 49.0 |
| 5 | 14440 | 11594 | 2846 | 733.6 | 3.925 | 39.3 |

Per-request data: `results.json`. Engine log lines from the start: `engine-log-excerpt.txt` (when this run started the server).

## Correctness and limitations

Not measured: this is a speed suite. The conversation uses the model's own replies, so token counts differ a little between runs. Machine details come from the server, `nvidia-smi`/`rocm-smi` and the OS; background load and power limits are not recorded. Total time of the suite: 183 s.