Speed sweep
qwen3-8-27b-gptq-4bit-rtx-3090-24gb-vllm-tp2-sweep
qwen3-8-27b-gptq-4bit-rtx-3090-24gb-vllm-tp2-sweepRecord
- Measured
- 2026-08-17T07:13:25.340Z
- Points
- 1
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status |
|---|---|---|---|---|---|
| 1 | 8,192 | 1,311.1 | 96.6 | 796.3 | observed |
Remaining fields
Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.
- id
- qwen3-8-27b-gptq-4bit-rtx-3090-24gb-vllm-tp2-sweep
- measured at
- 2026-08-17T07:13:25.340Z
- recipe id
- qwen3-8-27b-gptq-4bit-rtx-3090-24gb-vllm-tp2
- schema version
- local-ai-registry/v1
metrics
- concurrency
- 1
- latest point at
- 2026-08-17T07:13:25.340Z
- max context tokens
- 8,192
- peak generation tps
- 96.6
- peak prompt tps
- 1,311.1
- point count
- 1
rows
| concurrency | context tokens | decode tok s | decode tok s per stream | output tokens | peak vram gb | prefill tok s | samples |
|---|---|---|---|---|---|---|---|
| 1 | 8,192 | 96.6 | Unknown | 512 | 45.28 | 1,311.1 | 1 |
Provenance & metadata (1)
source
- kind
- leaderboard
- repository
- www.localmaxxing.com ↗