Meta-Llama-3.1-70B-Instruct
Meta developed and released the Meta Llama 3.1 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8B, 70B and 405B sizes.
text → text · made by Meta
Sold by 8 ways
| Seller | Lane | Rate |
|---|---|---|
| AWS Bedrock cloud | standard | $0.72per Mtok in$0.72per Mtok outmodels.dev · read 2026-11-01 |
| Oracle Cloud Infrastructure cloud | standard | $0.72per Mtok in$0.72per Mtok outraw.githubusercontent.com · read 2026-11-01 |
| Microsoft Azure AI cloud | standard | $2.68per Mtok in$3.54per Mtok outraw.githubusercontent.com · read 2026-11-01 |
| Hyperbolic aggregator | standard | $0.12per Mtok in$0.3per Mtok outraw.githubusercontent.com · read 2026-11-01 |
| Requesty aggregator | deepinfra | $0.23per Mtok in$0.4per Mtok out$0.23per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| DeepInfra aggregator | standard | $0.4per Mtok in$0.4per Mtok outapi.deepinfra.com · read 2026-08-25 |
| FriendliAI 0 ways | $0.6per Mtok in$0.6per Mtok out | |
| FriendliAI aggregatornot on the seller's list today | standard | $0.6per Mtok in$0.6per Mtok outraw.githubusercontent.com · read 2026-09-14 |
| Featherless AI aggregator | standard | $0.72per Mtok in$0.72per Mtok outapi.featherless.ai · read 2026-09-20 |
Measured 9 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 11thof 16 | The agent company — Epoch AI | % Score | 0.128 |
| 24thof 41 | Balrog — Epoch AI | Average progress | 0.279 |
| 27thof 217 | Mmlu — Epoch AI | EM | 0.801 |
| 77thof 108 | Math level 5 — Epoch AI | mean_score | 36.679 |
| 169thof 173 | Weirdml — Epoch AI | Accuracy | 8.97 |
| 177thof 197 | Lmca — Epoch AI | Score | 14.85 |
| 192ndof 235 | Dtbench — Epoch AI | Accuracy | 60 |
| 269thof 326 | GPQA Diamond | mean_score | 44.192 |
| 282ndof 306 | OTIS Mock AIME 2024-2025 | mean_score | 3.611 |
About
Meta-Llama-3.1-70B-Instruct — a text model from Meta, sold by 7 companies from $0.12 in and $0.3 out per million tokens, placed 27th of 217 on Mmlu — Epoch AI.
It takes text and returns text, with a context window of 32,768 tokens. It was published in July 2024, trained on material up to December 2023. Its sellers say it can call a tool. The catalogue files it under reasoning and agents. Seven companies sell it. The cheapest is $0.12 in and $0.3 out per million tokens at Hyperbolic. Beside the standard rate there is a separately routed lane. It has been measured on 9 boards, and stands best at 27th of 217 on Mmlu — Epoch AI.
Every current figure
- Maker
- Meta
- Register
- model
- Takes
- text
- Returns
- text
- Context
- 32,768 tokens
- Longest answer
- 16,384 tokens
- Published
- July 2024
- Knowledge to
- December 2023
- Parameters
- 70 billion · read from its own name
- Licence
- not read
- Sellers
- 7
- Maker's own price
- not read
- Price
- $0.12 in and $0.3 out per million tokens — Hyperbolic
- Boards
- 9
- Best place
- 27th of 217 — Mmlu — Epoch AI
Known as 7 names
meta-llama/Meta-Llama-3.1-70B-InstructNousResearch/Meta-Llama-3.1-70B-InstructOpenPipe/Llama-3.1-70B-InstructSillyTilly/Meta-Llama-3.1-70B-Instructunsloth/Meta-Llama-3.1-70B-Instructdeepinfra/meta-llama/Meta-Llama-3.1-70B-InstructMeta: Llama 3.1 70B Instruct