Llama 4 Scout vs Llama Nemotron Ultra 253B
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
|
Dby DeepInfra
3 providers:
Groq $0.110/$0.340
Meta $0.110/$0.340
DeepInfra $0.100/$0.300
|
Nby NVIDIA
|
|
|---|---|---|
| Overview | ||
| Status | Current open weights Open weights | Current open weights Open weights |
| Released | Apr 6, 2025 | Jan 1, 2025 |
| Pricing per million tokens | ||
| Input | $0.100/Mtok | $0.600/Mtok |
| Output | $0.300/Mtok | $3.60/Mtok |
| Blended avg | $0.200/Mtok | $2.10/Mtok |
| Specifications | ||
| Context window | 10M tokens | 128K tokens |
| Parameters | 17B (16 experts) | 253B |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
textimage
|
text
|
| Benchmarks sources: Vellum LLM Leaderboard (Jun 2026), Meta model card / Vellum LLM Leaderboard | ||
| Avg benchmark score | 46.1 | 70 |
| Perf / dollar | 230.5 | 33.3 |
| GPQA Diamond | 50 | 76 |
| SWE-Bench Verified | 28 | 30 vendor |
| LiveCodeBench | 32.8 | 64 |
| Humanity's Last Exam | 12 | 15 vendor |
| ARC-AGI 2 | 15 | 14 vendor |
| AIME 2025 | 45 | 45 vendor |
| MMMLU | 75 | 75 vendor |
| BFCL | 58 | 58 vendor |
| HumanEval | 80 | 80 vendor |
| MATH 500 | 65 | 65 vendor |
| Independent composite scores each on its own scale — not part of the average above | ||
| LMSYS Chatbot Arena | — | 1347.5 ELO |
| Providers | ||
| Available from |
Groq — $0.110/$0.340/Mtok
Meta — $0.110/$0.340/Mtok
DeepInfra — $0.100/$0.300/Mtok
|
NVIDIA — $0.600/$3.60/Mtok
|
Cost at scale — 1M tokens (50/50 input/output)
| Volume | Llama 4 Scout | Llama Nemotron Ultra 253B | Savings |
|---|---|---|---|
| 1M tokens | $0.2 | $2.1 | $1.9 (90.5%) |
| 10M tokens | $2 | $21 | $19 (90.5%) |
| 100M tokens | $20 | $210 | $190 (90.5%) |
| 1000M tokens | $200 | $2100 | $1900 (90.5%) |
Summary
Llama 4 Scout by DeepInfra costs $0.100/Mtok input and $0.300/Mtok output, with a 10M-token context window. It supports text, image input and is available from 3 providers.
Llama Nemotron Ultra 253B by NVIDIA costs $0.600/Mtok input and $3.60/Mtok output, with a 128K-token context window. It supports text input.
On a blended cost basis, Llama 4 Scout is 90.5% cheaper than Llama Nemotron Ultra 253B. It also has a larger context window.
On benchmarks, Llama Nemotron Ultra 253B scores higher (70 vs 46.1) on average. In terms of value, Llama 4 Scout has better performance per dollar (230.5 vs 33.3).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.