Llama 3 70B Instruct

Meta · Active · available in 5 of 31 regions

Streaming Tool useVisionPrompt cachingBatch inferenceEmbeddingsFine-tuning
Context window
8K
8,192 tokens
Max output
Input / 1M
$2.65
us-east-1
Output / 1M
$3.50

What Llama 3 70B Instruct is good for

editorial

Llama 3 70B Instruct offers a 8K-token context window, at the premium end of this provider’s range at $2.65 per 1M input tokens. Callable in 5 of 31 regions.

Suits

  • Nothing specific stands out from the published capabilities.

Think twice if

  • Long context. Only 8K tokens — long documents will need chunking, and agentic histories will hit the ceiling.
  • Agentic loops. No prompt caching published, so every step re-pays full price for the prompt prefix. Expensive for agents that resend a long context.
  • Function calling. No tool use, so it cannot drive an agent loop or call your APIs — it answers from what is in the prompt.
  • Documents with layout. Text only. Scanned PDFs, screenshots and charts need a vision model or an OCR step first.
  • High volume. At $2.65 per 1M input tokens it sits at the expensive end of the Meta range. Worth it for hard tasks, wasteful for bulk extraction.

This section is judgement, not data from AWS. It is composed from the capabilities, context window, price position and region coverage shown elsewhere on this page — so it stays in step with the daily snapshot rather than going stale.

Model & inference profile IDs

Base model ID — direct on-demand invoke

Pricing by region

USD per 1M tokens
Region Input Output Cache read Cache write Batch in Batch out Source
us-east-1 US East (N. Virginia) $2.65 $3.50 API
us-west-2 US West (Oregon) $2.65 $3.50 API
ca-central-1 Canada (Central) $3.05 $4.03 API
eu-west-2 EU (London) $3.45 $4.55 API
ap-south-1 Asia Pacific (Mumbai) $3.18 $4.20 API

Region availability

TPM / RPM are default account quotas from AWS Service Quotas where a model-specific limit is published. They are per-account defaults and adjustable on request.

Where this model actually runs

Under a strict residency constraint, Llama 3 70B Instruct can be served without the request leaving 4 of the 25 jurisdictions with a Bedrock region.

Common questions

How much does Llama 3 70B Instruct cost on Amazon Bedrock?
$2.65 per 1M input tokens and $3.50 per 1M output tokens in us-east-1.
Which AWS regions support Llama 3 70B Instruct?
5 of 31 regions: us-east-1, us-west-2, ca-central-1, eu-west-2, ap-south-1.
What is the context window of Llama 3 70B Instruct?
8,192 tokens (8K).
What is the model ID for Llama 3 70B Instruct on Bedrock?
meta.llama3-70b-instruct-v1:0.

Other Meta models