Speed sweep
qwen3-6-27b-dflash-q4-0-intel-arc-pro-b70-32gb-llama-cpp-tp4-sweep
qwen3-6-27b-dflash-q4-0-intel-arc-pro-b70-32gb-llama-cpp-tp4-sweepRecord
- Measured
- 2026-05-04T10:36:32.796Z
- Points
- 1
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status |
|---|---|---|---|---|---|
| 1 | 512 | — | 31.5 | — | 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-dflash-q4-0-intel-arc-pro-b70-32gb-llama-cpp-tp4-sweep
- measured at
- 2026-05-04T10:36:32.796Z
- recipe id
- qwen3-6-27b-dflash-q4-0-intel-arc-pro-b70-32gb-llama-cpp-tp4
- schema version
- local-ai-registry/v1
metrics
- concurrency
- 1
- inference engine version
- db44417+sycl-async-cpy+qwen-split+single-kernel-allreduce
- latest point at
- 2026-05-04T10:36:32.796Z
- max context tokens
- 512
- peak generation tps
- 31.482
- 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 | 512 | 31.482 | 31.482 | 512 | Unknown | Unknown | 1 |
Provenance & metadata (1)
source
- kind
- leaderboard
- repository
- www.localmaxxing.com ↗