NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 vs GPT-5.1-Codex

API pricing, benchmark indices and specs — updated 19 October 2026.

nvidia · A · Open weights

Params

560.5 B

Context

262,144

30d

1.1 M

openai · B · API-only

Input / 1M

$1.25

Output / 1M

$10.00

AA index

23.7

proprietary · 400K context

Specification comparison

Differences are highlighted; identical values are muted.

Specification	NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16	GPT-5.1-Codex
Deployment	Open weights · also via API	API-only (closed weights)
Parameters	560.5 B	n/a — API-only
Architecture	NemotronHForCausalLM	n/a — API-only
Context length	262,144	400,000
Licence	other	Proprietary
Commercial use	Unknown	Per API terms
Languages	12	n/a — API-only
Downloads 30d	1,067,855	n/a — API-only
Downloads all time	2.1 M	n/a — API-only
Quantizations on the Hub	SAFETENSORS	n/a — API-only
Gated	No	n/a — API-only
First seen on the Hub	2026-06-03	n/a — API-only
API input price / 1M tokens	$0.60	$1.25
API output price / 1M tokens	$2.40	$10.00
Est. cost per 1M input + 1M output tokens	$3.00	$11.25
Cached input / 1M tokens	$0.12	$0.13
AA Intelligence Index	22.9	23.7
AA Coding Index	49.3	—
AA Agentic Index	20.1	—
Max output tokens	182,520	128,000
API providers	3	1
Tool calling	Yes	Yes
Reasoning mode	Yes	Yes

Estimated cost of 1M input + 1M output tokens via OpenRouter

NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 $3.00

GPT-5.1-Codex $11.25

NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 is 3.8× cheaper at list price.

On RTX 4090 · 24 GB · 8K context unless noted

NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 · fp16 @ 256K ctx 3,924.2 GB / 24 GB

❌ Won’t fit