Llama 3 70B Instruct on Replicate API

Llama 3 · AI at Meta

ServerlessOpen Weights

Last refreshed 2026-07-09. Next refresh: weekly.

Why use Llama 3 70B Instruct on Replicate API?

Replicate API offers Llama 3 70B Instruct with pay-as-you-go pricing at $0.65/1M input tokens. Replicate is a cloud-based platform that enables users to run machine learning models easily and efficiently.

Compare Llama 3 70B Instruct across 18 providers to find the best fit for your use case
Input / 1M
$0.65
Output / 1M
$2.75
Cache
Not sourced
Batch
Not sourced

Setup recipe

Python + curl
Install
pip install replicate
Auth
export REPLICATE_API_TOKEN=...
Call
import replicate
output = replicate.run(
    "meta/meta-llama-3-70b-instruct",
    input={"prompt": "Hello"}
Model ID
meta/meta-llama-3-70b-instruct

Request example

import replicate

# reads REPLICATE_API_TOKEN from env
# meta/meta-llama-3-70b-instruct format: "owner/model-name" (latest version) or "owner/model-name:version-hash"
output = replicate.run(
    "meta/meta-llama-3-70b-instruct",
    input={"prompt": "Hello"}
)
# Output is a list or generator depending on the model
print("".join(output))

Gotchas

  • Use provider model ID "meta/meta-llama-3-70b-instruct", not the LLMReference slug "llama3-70b-instruct".
  • Replicate uses "owner/model-name" format (e.g. "meta/meta-llama-3-8b-instruct") for the latest version, or "owner/model-name:version-sha" to pin to a specific version. The REST endpoint splits owner and model-name into the path: /v1/models/{owner}/{model-name}/predictions.
  • The examples expect REPLICATE_API_TOKEN; rename it only if your application config maps the new variable.

Compare Llama 3 70B Instruct Across Providers

ProviderInput (per 1M)Output (per 1M)
GCP Vertex AI$1.20$3.60
AWS Bedrock$0.99$0.99
Microsoft Foundry$3.78$11.34
NVIDIA NIM——
DeepInfra$0.45$0.65
View all 18 providers →

Pricing

TypePrice (per 1M)
Input tokens$0.65
Output tokens$2.75

Capabilities

Structured Outputs

About Llama 3 70B Instruct

The Llama 3 70B Instruct model is a large language model with 70 billion parameters, released by Meta on April 18, 2024. It's an instruction-tuned variant optimized for conversational applications, utilizing an advanced auto-regressive transformer architecture. The model excels in following instructions and engaging in dialogue, having been trained on over 15 trillion tokens with a December 2023 knowledge cutoff. It demonstrates superior performance on industry benchmarks, scoring 82.0 on the MMLU (5-shot) test. The model incorporates extensive safety measures and optimizations, including RLHF, to enhance helpfulness and reduce harmful content generation.

Get Started