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. | 671B | 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. | 128k | 1M |
| 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 | ||
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 85.7% | 86.7% |
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% |
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% |
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 88.7% | — |
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 · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 30.6% | — |
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% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 96% | — |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 48 |
| Overview | ||
| Company | DeepSeek | NVIDIA |
| Release date | Dec 1 2025 | Jun 4 2026 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek-V3.2-SpecialevsClaude Fable 5.2Nemotron 3 UltravsClaude Fable 5.2DeepSeek-V3.2-SpecialevsGPT-6.2 AstraNemotron 3 UltravsGPT-6.2 AstraDeepSeek-V3.2-SpecialevsGemini 4 HydrogenNemotron 3 UltravsGemini 4 HydrogenDeepSeek-V3.2-SpecialevsMuse EmberNemotron 3 UltravsMuse EmberDeepSeek-V3.2-SpecialevsGrok 4.7Nemotron 3 UltravsGrok 4.7DeepSeek-V3.2-SpecialevsMistral Large 4Nemotron 3 UltravsMistral Large 4Frequently asked questions
Nemotron 3 Ultra leads DeepSeek-V3.2-Speciale on 1 of the 1 benchmark they both report (GPQA Diamond). DeepSeek-V3.2-Speciale shipped 185 days before Nemotron 3 Ultra, so benchmark comparisons should account for the intervening progress.
DeepSeek-V3.2-Speciale has 671B parameters, while Nemotron 3 Ultra has 550B. Context windows are 128k (DeepSeek-V3.2-Speciale) vs 1M (Nemotron 3 Ultra).
On GPQA Diamond, Nemotron 3 Ultra leads at 86.7% vs DeepSeek-V3.2-Speciale at 85.7%.