Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 550B | — |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 1M | 400k |
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across every provider on OpenRouter, the lab included. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.40DeepInfra | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $1.76DeepInfra | — |
| $210KT | — | |
| Benchmarks | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 49% | 45% |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 71.9% | — |
Next.js EvalsNext.js coding — Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better. | — | 46% |
Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. | 56.4% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 44.4% | — |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | 26.7% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 86.7% | — |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | 48 | — |
| Overview | ||
| Company | NVIDIA | OpenAI |
| Release date | Jun 4 2026 | Dec 18 2025 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Nemotron 3 UltravsClaude Fable 5.2GPT-5.2-CodexvsClaude Fable 5.2Nemotron 3 UltravsGemini 4 CarbonGPT-5.2-CodexvsGemini 4 CarbonNemotron 3 UltravsMuse EmberGPT-5.2-CodexvsMuse EmberNemotron 3 UltravsGrok 4.7GPT-5.2-CodexvsGrok 4.7Nemotron 3 UltravsDeepSeek-V4.1-Flash-1017GPT-5.2-CodexvsDeepSeek-V4.1-Flash-1017Nemotron 3 UltravsMistral Large 4GPT-5.2-CodexvsMistral Large 4Frequently asked questions
Nemotron 3 Ultra leads GPT-5.2-Codex on 1 of the 1 benchmark they both report (BullshitBench v2). GPT-5.2-Codex shipped 168 days before Nemotron 3 Ultra, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Nemotron 3 Ultra) vs 400k (GPT-5.2-Codex). Nemotron 3 Ultra is open weight, while GPT-5.2-Codex is closed.
On BullshitBench v2, Nemotron 3 Ultra leads at 49% vs GPT-5.2-Codex at 45%.