impulse

Llama3.2 3B Instruct

Open weights

Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.

API id llama3.2-3b-instructReleased 25 Sep 2024 (Sep 2024)
Input / 1M
$0.015
at Lambda AI
Output / 1M
$0.025
at Lambda AI
Context window
131K
tokens
Max output
131K
tokens per reply
Hosts
2
4 listings
Released
Sep 2024
25 Sep 2024

Pricing

List prices at Lambda AI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.015$0.025

What it costs a month

At Lambda AI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$6.38
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$8.10
  • Coding agent1,000 requests a day, 40K in / 2K out$19.50
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$42.00
Use your own numbers

Where to run it

2 hosts serve Llama3.2 3B Instruct. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
Lambda AIInference
llama3.2-3b-instruct
$0.015 / $0.025
meta.llama3-2-3b-instruct-v1:0
$0.15 / $0.15
+757% vs headline
Show 2 more listings: regions, snapshots and variants
Amazon Bedrock
us
us.meta.llama3-2-3b-instruct-v1:0
$0.15 / $0.15
+757% vs headline
eu
eu.meta.llama3-2-3b-instruct-v1:0
$0.19 / $0.19
+986% 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
Features
Parallel tool calls
Open weights
License: Llama Community License
Runs in about 7.7 GB of GPU memory
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 Lambda AI.

Smaller Meta models

No cheaper Meta model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

No cheaper open-weight match.

Benchmarks

Scores published on the model card.

  • BBH24.1%
  • GPQA3.8%
  • IFEval73.9%
  • MATH Lvl 517.7%
  • MMLU-Pro24.4%
  • MUSR1.4%
  • Open LLM Average24.2%

Model line

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

  1. Sep 2024Llama 3.2 3B InstructLatest$0.02 / $0.02
  2. Sep 2024Llama3.2 3B InstructThis page$0.015 / $0.025

Call it

Using Lambda AI's API.

Python
from litellm import completion

response = completion(
    model="lambda_ai/llama3.2-3b-instruct",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Llama3.2 3B Instruct cost?

On Lambda AI, Llama3.2 3B Instruct costs $0.015 per million input tokens and $0.025 per million output tokens.

Which APIs offer Llama3.2 3B Instruct?

Llama3.2 3B Instruct runs on 2 APIs: Lambda AI and Amazon Bedrock.

What is Llama3.2 3B Instruct's context window?

131K tokens, with up to 131K tokens in one reply.

Is Llama3.2 3B Instruct being retired?

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

Is there a newer version of Llama3.2 3B Instruct?

Yes. Llama 3.2 3B Instruct is the newest model in this line, released on 25 Sep 2024.

When was Llama3.2 3B Instruct released?

Meta released it on 25 Sep 2024.

What is Llama3.2 3B Instruct's API model id?

Use llama3.2-3b-instruct with Lambda AI. Other hosts use their own ids: meta.llama3-2-3b-instruct-v1:0 on Amazon Bedrock.

Sources

Raw catalog entry for llama3.2-3b-instruct on Lambda AI
{
  "input_cost_per_token": 1.5e-8,
  "litellm_provider": "lambda_ai",
  "max_input_tokens": 131072,
  "max_output_tokens": 131072,
  "mode": "chat",
  "output_cost_per_token": 2.5e-8,
  "supports_function_calling": true,
  "supports_parallel_function_calling": true,
  "supports_system_messages": true,
  "supports_tool_choice": true
}