Community benchmark on rtx-3050-plus-quadro-p6000-qwen3-8-flash-next-q2-0

Chiusa Adatta ai principianti
  • Qualcosa non va?
#2,214 0 commenti 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 3 giorni

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
1/5
Tempo stimato
Meno di un'ora
Idoneità per principianti
90/100
Tipo di issue
Documentazione
Chiarezza
Specificata chiaramente
Stato di attività
Attiva
Stack tecnologico
cpp, linux
Ambito
documentation

Direzione di ricerca

Questa issue contribuisce con risultati dei benchmark e dettagli sull’ambiente; inizia leggendo la documentazione dei benchmark del repository e il README.md collegato, quindi confronta il report con il formato dei benchmark usato da tools/strata_bench.py. Il lavoro è completato quando i risultati di RTX 3050 e Quadro P6000, il metodo di misurazione e i limiti sono registrati nella posizione documentata del progetto.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

Community benchmark on NVIDIA GeForce RTX 3050 + Quadro P6000

Measured on 2026-10-12 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 3050 + Quadro P6000, 30 GiB VRAM
  • CPU: Intel(R) Xeon(R) CPU X5670 @ 2.93GHz (12 cores, 24 threads); installed RAM: 94.4 GiB; storage: not measured
  • OS: Linux 6.8.0-146-generic; GPU driver: 580.178.04; CUDA: 13.0
  • Strata: project 0.1.43.1, commit a0116b27; engine 0.1.43.1
  • Background workloads and power limits: not recorded

Model and configuration

  • Model and quantization: qwen3.8-flash-next-q2_0
  • Context: 131072 tokens; parallel slots: 1
  • Settings: the install's own configuration, unchanged; thinking off for the suite; temperature 0
strata --pack q2_0 --native Qwen3.8-Flash-Next-GSQ-RCO-Q2_0-00001-of-00002.gguf --ple-gguf Qwen3.8-Flash-Next-GSQ-RCO-Q2_0-00002-of-00002.gguf --expert-profile expert-profile.bin --expert-cache auto --prefill auto --spec 4 --mtp rt --max-context 131072 --kv int8 --kv-resident 32768 --vision --vram-reserve-mib 700 --control-vector-scaled Qwen3.8-Flash-Next-experimental-speed-projection.gguf:1.0 --control-vector-layer-range 4 44 --cvec-mode project --cvec-dir per-layer --pcie-frac 0.55 --spec-min-p 0.70 --adapt-every 1 --adapt-swaps 80 --adapt-decay 0.97

Method

Run by python tools/strata_bench.py against the HTTP API of a Strata server (started by the benchmark from the install configuration); 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 33.4 (27.1-40.7) 1.06 (0.90-1.60)
4K prompt, fresh 4019 (4019-4019) 0 (0-0) 24 cap 3 253.6 (252.7-256.1) not measured 15.90 (15.74-15.95)
16K prompt, fresh 16058 (16058-16058) 0 (0-0) 24 cap 3 244.8 (242.3-249.0) not measured 65.65 (64.57-66.34)
agent session, follow-up turns 14464 (14416-14475) at the last turn see below up to 64 3 see below 33.8 (22.1-39.4) 17.67 (9.14-21.18)

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: 68.6 % (68.6-68.8); earlier-turn tokens read again per session: 74 (4-74) (earlier turns that the cache did not hold; the 13633 (13633-13633) new tokens of the files are read once); first turn (nothing cached) first token: 5.22 s (5.16-6.45).

Turn (run 1) Prompt tokens Reused Read Prompt tok/s First token (s) Decode tok/s
0 609 0 609 95.2 6.45 18.7
1 3701 621 3080 163.0 18.963 28.2
2 5234 3764 1470 135.8 10.889 34.7
3 8678 5227 3451 163.5 21.18 31.8
4 11584 8724 2860 162.3 17.69 33.8
5 14464 11618 2846 160.0 17.855 22.1

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: 589 s.

engine-log-excerpt.txt
README.md
results.json

Lingua principale
C++
Stelle
11.6k
Fork
1k
Nessuna PR unita negli ultimi 30g

Preparare l'ambiente

Questo progetto non fornisce container di sviluppo, Dockerfile né guida per i contributori, quindi l'ambiente è a tuo carico: parti dal suo README e consulta la nostra guida al primo contributo per i passaggi generali.

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di Niko1221/Strata

Tutte le issue di Niko1221/Strata

Issue simili

Altre issue su C++

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.