Speed sweep
qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr-sweep
qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr-sweepRecord
- Measured
- 2026-06-18T19:40:05.958Z
- Points
- 1
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status |
|---|---|---|---|---|---|
| 1 | 8,192 | 664.9 | 50 | 1,452.9 | observed |
Remaining fields
Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.
- id
- qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr-sweep
- measured at
- 2026-06-18T19:40:05.958Z
- recipe id
- qwen3-6-27b-q4-k-m-rx-7900-xtx-24gb-llama-cpp-tp1-70ju7hgr
- schema version
- local-ai-registry/v1
metrics
- concurrency
- 1
- latest point at
- 2026-06-18T19:40:05.958Z
- max context tokens
- 8,192
- peak generation tps
- 49.96
- peak prompt tps
- 664.87
- 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 | 49.96 | Unknown | 512 | 22.9 | 664.87 | 1 |
Provenance & metadata (1)
source
- kind
- leaderboard
- repository
- www.localmaxxing.com ↗