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. | 1B/4B/12B/27B | 253B |
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 | 128k |
| 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. | $0.04DeepInfra | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.10DeepInfra | — |
| $15.00DeepInfra · bf16 | — | |
| Benchmarks | ||
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. | 29.7% | 66.3% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 42.4% | 76% |
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. | 3% | — |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | 87.8% | — |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | 95.9% | — |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 64.9% | — |
| Overview | ||
| Company | NVIDIA | |
| Release date | Mar 12 2025 | Apr 8 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Gemma 3vsClaude Haiku 5.5Llama Nemotron Ultra 253BvsClaude Haiku 5.5Gemma 3vsGPT-6.1 SolLlama Nemotron Ultra 253BvsGPT-6.1 SolGemma 3vsMuse Spark OpenLlama Nemotron Ultra 253BvsMuse Spark OpenGemma 3vsGrok 4.7Llama Nemotron Ultra 253BvsGrok 4.7Gemma 3vsDeepSeek-V4.1-Flash-1017Llama Nemotron Ultra 253BvsDeepSeek-V4.1-Flash-1017Gemma 3vsMistral Large 4Llama Nemotron Ultra 253BvsMistral Large 4Frequently asked questions
Llama Nemotron Ultra 253B leads Gemma 3 on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). Gemma 3 shipped 27 days before Llama Nemotron Ultra 253B, so benchmark comparisons should account for the intervening progress.
Gemma 3 has 1B/4B/12B/27B parameters, while Llama Nemotron Ultra 253B has 253B. Context windows are 128k (Gemma 3) vs 128k (Llama Nemotron Ultra 253B).
On LiveCodeBench, Llama Nemotron Ultra 253B leads at 66.3% vs Gemma 3 at 29.7%. On GPQA Diamond, Llama Nemotron Ultra 253B leads at 76% vs Gemma 3 at 42.4%.