Llama 3.1-70B

Meta AIOpen weights

Llama 3.1-70B by Meta AI costs — per 1M input tokens and — per 1M output tokens, with a —-token context window.

Pricing

Input (per 1M tokens)

—

Output (per 1M tokens)

—

Cached input (per 1M)

—

Specifications

Provider: Meta AI
Context window: —
Parameters: 70B
Released: Jul 2024
Open weights: Yes
Frontier model: No

Benchmarks

Epoch Capabilities Index: 125.9

Speed & latency

Time to first token

0.28s

Output speed

113 tok/s

Fastest host: bedrock (via vercel-ai-gateway).

Compare Llama 3.1-70B with…

Llama 3.1-70B vs Pareto Latest$7.50/1M Llama 3.1-70B vs Llama 3.3 70B Versatile$0.79/1M Llama 3.1-70B vs Claude Opus 5$25.00/1M Llama 3.1-70B vs MiMo-V2-Pro$3.00/1M Llama 3.1-70B vs GPT-4o mini$0.60/1M Llama 3.1-70B vs o3-pro$40.00/1M

FAQ

How much does Llama 3.1-70B cost?

Llama 3.1-70B is priced at — per 1M input tokens and — per 1M output tokens. Output tokens cost more than input.

What is Llama 3.1-70B's context window?

Llama 3.1-70B supports a context window of — tokens — the maximum prompt plus response length it can handle in a single request.

Is Llama 3.1-70B open weight?

Yes — Llama 3.1-70B has openly available weights.

Data as of 2026-10-14. Source: Epoch AI, Data on AI Models (CC BY 4.0), epoch.ai. Pricing is per 1M tokens (USD); confirm with the provider before production use.