impulse

Llama3 8B Instruct

Open weights
API id meta.llama3-8b-instruct-v1:0
A newer model in this line is out: Llama 3.3 8B Instruct.
Amazon Bedrock (except US government regions) retires it on 30 Mar 2027 (in 132 days); Amazon Bedrock (US government regions only) on 9 May 2027. 3 hosts keep serving it.
Input / 1M
$0.30
at Amazon Bedrock
Output / 1M
$0.60
at Amazon Bedrock
Context window
8K
tokens
Max output
8K
tokens per reply
Hosts
3
12 listings
Released
-
date not published

Pricing

List prices at Amazon Bedrock, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.30$0.60

What it costs a month

At Amazon Bedrock list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$135
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$166
  • Coding agent1,000 requests a day, 40K in / 2K out$396
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$864
Use your own numbers

Where to run it

3 hosts serve Llama3 8B Instruct. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
llama3-8b-instruct
$0.20 / $0.20
−47% vs headline
meta.llama3-8b-instruct-v1:0
$0.30 / $0.60
Retires 30 Mar 2027
llama3-8b-instruct-maas
-
Show 9 more listings: regions, snapshots and variants
Amazon Bedrock
us-east-1
meta.llama3-8b-instruct-v1:0
$0.30 / $0.60
Retires 30 Mar 2027
us-gov-west-1
meta.llama3-8b-instruct-v1:0
$0.30 / $0.60
Retires 9 May 2027
us-west-1
meta.llama3-8b-instruct-v1:0
$0.30 / $0.60
Retires 30 Mar 2027
eu-west-1
meta.llama3-8b-instruct-v1:0
$0.32 / $0.65
+7% vs headline
Retires 30 Mar 2027
ca-central-1
meta.llama3-8b-instruct-v1:0
$0.35 / $0.69
+16% vs headline
Retires 30 Mar 2027
ap-south-1
meta.llama3-8b-instruct-v1:0
$0.36 / $0.72
+20% vs headline
Retires 30 Mar 2027
eu-west-2
meta.llama3-8b-instruct-v1:0
$0.39 / $0.78
+30% vs headline
Retires 30 Mar 2027
sa-east-1
meta.llama3-8b-instruct-v1:0
$0.50 / $1.01
+67% vs headline
Retires 30 Mar 2027
us-gov-east-1
meta.llama3-8b-instruct-v1:0
$0.30 / $2.65
+137% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: Llama Community License
Runtimes: Ollama, TGI, llama.cpp, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

  • GradientAI
    llama3-8b-instruct
    $0.20 / $0.20
    47% less

Smaller Meta models

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    47% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    47% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    30% less

Open-weight picks

Model line

Newer and older versions, at each lab's own price.

  1. UndatedLlama 3.3 8B InstructLatestNot listed
  2. Jul 2024Llama 3.1 8B Instruct$0.02 / $0.05
  3. Jul 2024Llama3.1 8B Instruct$0.025 / $0.04
  4. UndatedLlama 3 8B Instruct$0.04 / $0.04
  5. UndatedLlama3 8B InstructThis page$0.30 / $0.60

Call it

Using Amazon Bedrock's API.

Python
from litellm import completion

response = completion(
    model="bedrock/meta.llama3-8b-instruct-v1:0",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Llama3 8B Instruct cost?

On Amazon Bedrock, Llama3 8B Instruct costs $0.30 per million input tokens and $0.60 per million output tokens. The cheapest host is GradientAI, at $0.20 per million input tokens and $0.20 per million output tokens.

Which APIs offer Llama3 8B Instruct?

Llama3 8B Instruct runs on 3 APIs: GradientAI, Amazon Bedrock and Vertex AI.

What is Llama3 8B Instruct's context window?

8K tokens, with up to 8K tokens in one reply.

Is Llama3 8B Instruct being retired?

Amazon Bedrock (except US government regions) retires it on 30 Mar 2027; Amazon Bedrock (US government regions only) on 9 May 2027. 3 hosts keep serving it.

Is there a newer version of Llama3 8B Instruct?

Yes. Llama 3.3 8B Instruct is the newest model in this line.

What are cheaper alternatives to Llama3 8B Instruct?

Models that cost less and keep most of what it can do: Muse Spark 1.3 Contributor on Meta Model API (67% less), Claude Haiku 5.5 on Anthropic (47% less), GPT-6 Luna on OpenAI (47% less) and Qwen3.5 397B A17B on AIHubMix (1% less). Llama3 8B Instruct itself costs 47% less on GradientAI. Savings are on a mix of 3 input tokens to 1 output token.

What is Llama3 8B Instruct's API model id?

Use meta.llama3-8b-instruct-v1:0 with Amazon Bedrock. Other hosts use their own ids: llama3-8b-instruct on GradientAI, llama3-8b-instruct-maas on Vertex AI.

Sources

Raw catalog entry for meta.llama3-8b-instruct-v1:0 on Amazon Bedrock
{
  "deprecation_date": "2027-03-30",
  "deprecation_source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-meta-llama-3-8b-instruct.md",
  "input_cost_per_token": 3e-7,
  "litellm_provider": "bedrock",
  "max_input_tokens": 8192,
  "max_output_tokens": 8192,
  "mode": "chat",
  "output_cost_per_token": 6e-7,
  "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-meta-llama-3-8b-instruct.md"
}