Speed sweep
qwen3-8-27b-q4-0-rocmfp4-strix-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1-sweep
qwen3-8-27b-q4-0-rocmfp4-strix-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1-sweepRecord
- Measured
- 2026-08-15T01:59:30.020Z
- Points
- 1
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status |
|---|---|---|---|---|---|
| 1 | 65,536 | 300 | 23.9 | 27,469.3 | 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-q4-0-rocmfp4-strix-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1-sweep
- measured at
- 2026-08-15T01:59:30.020Z
- recipe id
- qwen3-8-27b-q4-0-rocmfp4-strix-ryzen-ai-max-plus-395-128gb-llama-cpp-tp1
- schema version
- local-ai-registry/v1
metrics
- concurrency
- 1
- inference engine version
- ROCmFPX b9219 / 1faa48eef / ROCm 7.2.4
- latest point at
- 2026-08-15T01:59:30.020Z
- max context tokens
- 65,536
- peak generation tps
- 23.88
- peak prompt tps
- 300.03
- 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 | 65,536 | 23.88 | 23.88 | 256 | Unknown | 300.03 | 1 |
Provenance & metadata (1)
source
- kind
- leaderboard
- repository
- www.localmaxxing.com ↗