The Open-Source AI Stack
RSS
All models

Models · Compare

OpenAI o4-mini vs Llama-3.1-Nemotron Ultra 253B v1

Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.

Specs

Field A: OpenAI o4-mini B: Llama-3.1-Nemotron Ultra 253B v1
Released 2025-04-162025-04-11
Developer OpenAINVIDIA
Openness ProprietaryOpen weights
License ProprietaryNVIDIA Open Model License
OSI-approved nono
Data released noyes
Training code nono
Architecture unknowndense
Total params 253B
Active params
Experts
Context window 200K131K
Attention unknownskip-attention
Position enc. unknownrope
Pretraining tokens 65B
Post-training rlhfsft, grpo
Training hardware H100
$/M input $1.10
$/M output $4.40
Output tok/sec 151.2

Benchmarks

Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.

General reasoning

MMLU-Pro 83.2 2026-05-21
GPQA-Diamond 78.4 2026-05-21 76.0 2025-04-11

Code

LiveCodeBench 85.9 2026-05-21 66.3 2025-04-11

Math

MATH 98.9 2026-05-21 97.0 2025-04-11
AIME 2024 94.0 2026-05-21
AIME 2025 90.7 2026-05-21 72.5 2025-04-11

Held-out / arena

IFEval 88.8 2025-04-11

Context · A

Released April 16, 2025 alongside o3 as a smaller multimodal reasoning model with a 200K-token context. Initially launched to paid subscribers and expanded to all ChatGPT users on April 24, 2025; OpenAI positioned it as the cost-efficient successor to o3-mini with image-input support.

Context · B

Top of NVIDIA's Llama-Nemotron family, distilled from Llama 3.1 405B via Neural Architecture Search with skip attention, variable FFN, and FFN fusion. Released April 11 2025; single-node 8x H100 BF16 inference, 4x H100 FP8. Post-trained through SFT and GRPO RL stages.

OpenAI o4-mini detail → · Llama-3.1-Nemotron Ultra 253B v1 detail →