Nox Orion 2 (Retired)

Retired flagship model · Released in Nox: September 23, 2026 · Page last updated: November 10, 2026

Retired — the first flagship of the Nox 2 generation

Orion 2 was the direct upgrade to 1.4 Orion and the first flagship of the Nox 2 generation. It kept everything 1.4 Orion could do — the same tools, thinking levels, and answer standard — while responding faster overall and costing less. It was an 824-billion-parameter mixture-of-experts model, 46 billion active per token, with a roughly one-million-token context window. On October 22, 2026 it was succeeded by Nox Orion 2.1, which runs the same engine with a new peer-review answer pipeline.

Architecture: mixture-of-experts, 824 billion parameters total, 46 billion active per token — the 700-billion-parameter backbone of 1.3 Astra extended with roughly 100 billion parameters of additional trainable components.

Length: Orion 2 sized each answer to the question — short for simple questions, up to 9,000 characters for multi-part ones — and covered the direct answer, main alternatives, next steps, and what would warrant urgent care.

Measured on the 1.4 Orion pipeline in September 2026: HealthBench Hard, Nox self-run, official open protocol, all 1,000 examples, scored 46.4% (±0.8) at Ultra and 53.3% (±1.0) at Hyper Ultra. Orion 2 ran the same engine and answer pipeline, and these results were never re-measured under the Orion 2 name.

Hyper Ultra was Orion 2’s sixth and deepest thinking level, in Plan mode on MAX: several passes over one question — draft, self-written rubric, self-grade, revision — about five to six minutes per answer.

Multimodal and tool-using: up to ten photos per message; live web access in Plan mode with up to ten search-and-read rounds per answer, every source cited; inline health and anatomy image generation; thinking time shown on every answer and a live step timeline in Plan mode.

Nox Orion 2 was retired on October 22, 2026, succeeded by Nox Orion 2.1 — the same engine with a peer-review answer pipeline built from the failure modes measured on HealthBench Professional.

Capabilities

• 824-billion-parameter mixture-of-experts model, 46 billion active per token
• ~1,000,000-token context window
• Deliberate internal reasoning pass before every answer, thinking time shown
• Six thinking levels, Light to Hyper Ultra
• Dynamic answer length, up to a 9,000-character cap
• Photo input — up to ten images per message
• Live web access in Plan mode — up to ten search-and-read rounds per answer, every source cited
• Image generation — health and anatomy illustrations inline

Model details

• Architecture — Mixture-of-experts — 824 billion parameters total, 46 billion active per token
• Context window — ~1,000,000 tokens
• Answer length — Dynamic — sized to the question, up to a 9,000-character hard cap
• HealthBench Hard — Measured on the 1.4 Orion pipeline (same engine): 46.4% (±0.8) at Ultra, September 5, 2026; 53.3% (±1.0) at Hyper Ultra, September 11, 2026 — Nox's own self-runs, official open protocol, all 1,000 examples; not re-measured under the Orion 2 name; not independently verified
• Thinking levels — Six — Light, Balanced, Extra, Max, Ultra, Hyper Ultra
• Released — September 23, 2026
• Status — Retired on October 22, 2026 — succeeded by Nox Orion 2.1

Availability: Retired on October 22, 2026 — succeeded by Nox Orion 2.1.

Nox Orion 2 was retired on October 22, 2026. Succeeded by Nox Orion 2.1 (same engine, new peer-review answer pipeline). Current replacement: Nox Orion 2.1.

Nox Orion 2 — frequently asked questions

What is Nox Orion 2?

Nox Orion 2 was Nox's flagship model, retired on October 22, 2026. Orion 2 was the direct upgrade to 1.4 Orion and the first flagship of the Nox 2 generation. It kept everything 1.4 Orion could do — the same tools, thinking levels, and answer standard — while responding faster overall and costing less. It was an 824-billion-parameter mixture-of-experts model, 46 billion active per token, with a roughly one-million-token context window. On October 22, 2026 it was succeeded by Nox Orion 2.1, which runs the same engine with a new peer-review answer pipeline.

What could Nox Orion 2 do?

Nox Orion 2's capabilities: 824-billion-parameter mixture-of-experts model, 46 billion active per token; ~1,000,000-token context window; Deliberate internal reasoning pass before every answer, thinking time shown; Six thinking levels, Light to Hyper Ultra; Dynamic answer length, up to a 9,000-character cap; Photo input — up to ten images per message; Live web access in Plan mode — up to ten search-and-read rounds per answer, every source cited; Image generation — health and anatomy illustrations inline.

What architecture does Nox Orion 2 use?

Architecture: Mixture-of-experts — 824 billion parameters total, 46 billion active per token

How large is Nox Orion 2's context window?

Context window: ~1,000,000 tokens

How long are Nox Orion 2's answers?

Answer length: Dynamic — sized to the question, up to a 9,000-character hard cap

Which thinking levels did Nox Orion 2 offer?

Thinking levels: Six — Light, Balanced, Extra, Max, Ultra, Hyper Ultra

Is Nox Orion 2 still available?

No. Nox Orion 2 was retired on October 22, 2026 and is no longer part of the Nox lineup. Succeeded by Nox Orion 2.1 (same engine, new peer-review answer pipeline).

Related Nox models

• Nox Orion 2.1 — The Nox 2.1 flagship, with a peer-review answer pipeline
• Nox GHI 1 — The officially released model system that powers Pebs, Nox's agent
• Nox Orion 2.5 — The new flagship — workspace tools built in, health research tools, and a staged review at Ultra
• Nox Orion Flash 2.5 — The fast Orion 2.5 — native photo input, workspace tools built in, and the same review at Ultra
• Nox Orion 2.6 Cyber Preview — A dedicated security model for defensive and authorized security work — hardened against misuse and prompt injection
• Nox Orion 2.1 Pro — Visualize your body: 3D models, code, and MP4 video
• Nox Orion Flash 2.1 — The fast Orion on a new engine — 552B mixture-of-experts, native photo input
• Nox Vela 2.5 — The fast everyday model on every plan — now reads images
• Nox Voice 2 — The most powerful Nox voice yet — expressive, interruptible, Plan mode in the call

Back to the Nox model family · Plans and pricing · Nox features