The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...
Tested on 18 benchmarks · BenchGecko score 47.9. Top scores: Chatbot Arena Elo — Overall (1317.8%), IFEval (90.0%), MMLU (81.7%).
Gemma 4 26B A4B scores 47.8 (100% as good) at $0.07/1M input · 32% cheaper
Code editing benchmark from the Aider project. Measures ability to apply targeted code changes while maintaining correctness and style.
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.
Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.
HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.
Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
- Typetext
- Context131K tokens (~66 books)
- ReleasedDec 2024
- LicenseOpen Source
- StatusActive
- Cost / Message~$0.001
Frequently Asked Questions
Key facts · as of 2026-10-05
- Llama 3.3 70B Instruct by Meta. BenchGecko score 47.9, rank 157 of 312 scored models (normalized average of public benchmark scores).
- List price $0.10 input · $0.32 output per 1M tokens (as of 2026-10-05).
- Sold by 11 providers (as of 2026-10-05): DeepInfra (fp8) $0.10 in / $0.32 out · Novita (bf16) $0.14 in / $0.40 out · AkashML (fp8) $0.20 in / $0.52 out · Parasail (fp8) $0.22 in / $0.50 out · Cloudflare (fp8) $0.29 in / $2.25 out · and 6 more. Every provider
How to cite · data as of 2026-10-05
Llama 3.3 70B Instruct · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/llama-3-3-70b-instruct
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP