Measurement wanted · request DR-006 · opened 2026-10-31
Drex-Max v1 on a 16GB GPU: not measured yet
We don't have a Drex-Max v1 speed run on a 16GB GPU yet, so there is no tokens-per-second number here. People asked for it, and this page is where the result will appear.
Drex-Max is an answer-only decision model: it returns an answer for a given context and does not stream generated tokens, so there is no decode tok/s to measure. The llm-speed CLI cannot measure it yet; this page records the request until it can.
What's being requested
- Model
- Drex-Max v1 · nace-ai/drex-max-v1
- Hardware
- a 16GB GPU
- Also requested
- RTX 4090, RTX 3090, Apple M4
Runs we already have
No Drex-Max v1 runs on any hardware yet.
Get an alert
Email alerts for this page are switching on soon. Until then, follow @llmspeed on X, where new runs are posted.