Nox 1.3 Vela
The fast 1.3 model on every plan
1.3 Vela is the everyday model of the 1.3 family and the one available on every plan, including Free. It runs on a lean 21-billion-parameter mixture-of-experts engine that activates only about 3.6 billion parameters per token — a small active set that is exactly what makes it near-instant.
Vela is the newest generation of Nox's fast models and Light 1.1's successor. It answers with the 1.3 generation's cleaner organisation while keeping latency low enough that replies land in roughly the time the question takes to read.
The engine is a mixture-of-experts transformer on an open-weights foundation: 21 billion total parameters arranged as 32 experts per layer, with about 3.6 billion parameters active per token. It is a genuine reasoning model — it runs a brief thinking pass before answering — but the pass is short by design, so speed stays the headline property.
Its context window is 128,000 tokens, which keeps long conversations fully in view. Photo input is supported and handled automatically, and it carries the family's shared toolset: inline charts, connected-app actions, and consent-gated live web access with cited sources.
Like every Nox model, Vela sits behind Leo, the deterministic safety layer whose measured recall and false-positive rate are published from a generated evaluation and identical family-wide — the same gate in front of the fastest model as the flagship.
1.3 Vela is available on every plan, including Free. A conversation can be handed up to 1.3 Nova on Pro or MAX mid-thread, with its history kept intact.
Capabilities
- Near-instant replies from a ~3.6-billion-active-parameter mixture-of-experts engine
- Brief built-in reasoning pass before every answer
- 128,000-token context window
- Native photo understanding, inline charts, and connected-app actions
- Consent-gated live web access with every source cited
Model details
- Architecture — Mixture-of-experts on an open-weights foundation — 21 billion parameters total, 32 experts per layer, ~3.6 billion active per token
- Context window — 128,000 tokens
- Reasoning — Brief reasoning pass before each answer — thinks first, but only briefly
- Latency — The fastest 1.3 model — optimized for near-instant replies
- Photo understanding — Yes — attach photos as usual; Nox handles them automatically
- Live web access — Yes — consent-gated; Nox asks before searching, then cites every source it uses
- Safety layer — Runs behind Leo, Nox's deterministic red-flag detector, on every message
- Available on — Free, Pro, and MAX
Availability: Available on every plan, including Free.