impulse

Meta Llama 3 8B Instruct

Open weights
API id Meta-Llama-3-8B-Instruct
Input / 1M
$0.03
at DeepInfra
Output / 1M
$0.06
at DeepInfra
Context window
8K
tokens
Max output
8K
tokens per reply
Hosts
3
3 listings
Released
-
date not published

Pricing

List prices at DeepInfra, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.03$0.06

What it costs a month

At DeepInfra list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$13.50
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$16.56
  • Coding agent1,000 requests a day, 40K in / 2K out$39.60
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$86.40
Use your own numbers

Where to run it

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

HostIn / out per 1M
DeepInfraInference
Meta-Llama-3-8B-Instruct
$0.03 / $0.06
ReplicateInference
meta-llama-3-8b-instruct
$0.05 / $0.25
+167% vs headline
Meta-Llama-3-8B-Instruct
$0.15 / $0.15
+300% 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

No host sells it for less than DeepInfra.

Smaller Meta models

No cheaper Meta model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • Llama 3.1 8B
    Meta · on LlamaGate
    $0.03 / $0.05
    Has every key capability
    7% less
  • Mistral Nemo Instruct 2407
    Mistral AI · on DeepInfra
    $0.029 / $0.03
    Has every key capability
    22% less
  • gpt-oss-20b
    OpenAI · on Darkbloom
    $0.0145 / $0.07
    Has every key capability
    24% less
  • Qwen2.5 Coder 7B
    Alibaba Qwen · on Nebius AI Studio
    $0.01 / $0.03
    Has every key capability
    60% less

Model line

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

  1. UndatedMeta Llama 3.1 8B InstructLatest$0.02 / $0.06
  2. UndatedMeta Llama 3 8B InstructThis page$0.03 / $0.06

Call it

Using DeepInfra's API.

Python
from litellm import completion

response = completion(
    model="deepinfra/Meta-Llama-3-8B-Instruct",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Meta Llama 3 8B Instruct cost?

On DeepInfra, Meta Llama 3 8B Instruct costs $0.03 per million input tokens and $0.06 per million output tokens.

Which APIs offer Meta Llama 3 8B Instruct?

Meta Llama 3 8B Instruct runs on 3 APIs: DeepInfra, Replicate and Anyscale.

What is Meta Llama 3 8B Instruct's context window?

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

Is Meta Llama 3 8B Instruct being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Meta Llama 3 8B Instruct?

Yes. Meta Llama 3.1 8B Instruct is the newest model in this line.

What are cheaper alternatives to Meta Llama 3 8B Instruct?

Models that cost less and keep most of what it can do: Llama 3.1 8B on LlamaGate (7% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Meta Llama 3 8B Instruct's API model id?

Use Meta-Llama-3-8B-Instruct with DeepInfra. Other hosts use their own ids: meta-llama-3-8b-instruct on Replicate.

Sources

Raw catalog entry for Meta-Llama-3-8B-Instruct on DeepInfra
{
  "max_input_tokens": 8192,
  "max_output_tokens": 8192,
  "input_cost_per_token": 3e-8,
  "output_cost_per_token": 6e-8,
  "litellm_provider": "deepinfra",
  "mode": "chat",
  "supports_tool_choice": true,
  "supports_function_calling": true
}