Speed sweep
internvl3-14b-instruct-q4-k-m-rtx-3090-24gb-llama-cpp-tp1-sweep
internvl3-14b-instruct-q4-k-m-rtx-3090-24gb-llama-cpp-tp1-sweepRecord
- Measured
- 2026-07-19T22:12:56.590Z
- Points
- 1
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status |
|---|---|---|---|---|---|
| 1 | 512 | 2,251.3 | 52.4 | 347.9 | observed |
Remaining fields
Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.
- id
- internvl3-14b-instruct-q4-k-m-rtx-3090-24gb-llama-cpp-tp1-sweep
- measured at
- 2026-07-19T22:12:56.590Z
- recipe id
- internvl3-14b-instruct-q4-k-m-rtx-3090-24gb-llama-cpp-tp1
- schema version
- local-ai-registry/v1
metrics
- concurrency
- 1
- latest point at
- 2026-07-19T22:12:56.590Z
- max context tokens
- 512
- peak generation tps
- 52.39
- peak prompt tps
- 2,251.32
- 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 | 52.39 | 52.39 | 128 | Unknown | 2,251.32 | 1 |
Provenance & metadata (1)
source
- kind
- leaderboard
- repository
- www.localmaxxing.com ↗