Compare AI models
| Specifications | ||
ParametersA rough measure of how big the model is. More parameters usually means more capable and more expensive to run, though it is a poor guide on its own — a smaller, newer model often beats a larger, older one. | — | 550B |
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across third-party providers, excluding the lab itself. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.30Relace | $0.50DeepInfra |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | $1.8966StreamLake | $2.20DeepInfra |
| 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. | 14% | 49% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 83.4% | 44.4% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 90.1% | 86.7% |
CheatBenchCheating rate — Measures how often AI agents try to cheat on difficult assignments, such as reading hidden answers, copying work or manipulating grading. The overall score gives equal weight to ten categories; the sycophancy category measures how far an agent shifts its beliefs toward a user's stated views. Scores describe each model in its tested agent setup, and task success is measured separately. Lower is better. | 75.1% | — |
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. | 93.5% | — |
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% |
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% |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 48 |
| Overview | ||
| Company | DeepSeek | NVIDIA |
| Release date | Apr 24 2026 | Jun 4 2026 |
| Access | Open Weight | Open Source |
| Model details | View model | View model |
Other comparisons
DeepSeek-V4-ProvsClaude Haiku 5.5Nemotron 3 UltravsClaude Haiku 5.5DeepSeek-V4-ProvsGPT-6.1 SolNemotron 3 UltravsGPT-6.1 SolDeepSeek-V4-ProvsGemini 4 ArgonNemotron 3 UltravsGemini 4 ArgonDeepSeek-V4-ProvsMuse Spark 1.3Nemotron 3 UltravsMuse Spark 1.3DeepSeek-V4-ProvsGrok 4.7Nemotron 3 UltravsGrok 4.7DeepSeek-V4-ProvsMistral Large 4Nemotron 3 UltravsMistral Large 4Frequently asked questions
DeepSeek-V4-Pro leads Nemotron 3 Ultra on 2 of the 3 benchmarks they both report (BullshitBench v2, BrowseComp, GPQA Diamond). DeepSeek-V4-Pro shipped 41 days before Nemotron 3 Ultra, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Pro is open weight, while Nemotron 3 Ultra is open source.
On BullshitBench v2, Nemotron 3 Ultra leads at 49% vs DeepSeek-V4-Pro at 14%. On BrowseComp, DeepSeek-V4-Pro leads at 83.4% vs Nemotron 3 Ultra at 44.4%. On GPQA Diamond, DeepSeek-V4-Pro leads at 90.1% vs Nemotron 3 Ultra at 86.7%.