Adaptive multi-token prediction selects draft horizons online
On Llama and Qwen pairings, it was up to 4.1x faster than standard decoding and beat a fixed window in 8 of 10 settings.
46
AI model
Qwen3-4B-Inst-2507-MTP
- Mentions · 7d1
- Mentions · 30d1
- All mentions1
- First seenOct 20, 2026
- Last mentioned5d ago
Timeline
1 articles · newest firstAppears with
- Llama-3.1-8B-Instructai model · 1
- Metaorganization · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.