impulse

Llama 3.2 3B Instruct Fast

Open weights
API id llama-3.2-3b-instruct-fast
Input / 1M
$0.05
at ZeroGPU
Output / 1M
$0.30
at ZeroGPU
Context window
131K
tokens
Max output
8K
tokens per reply
Hosts
1
1 listings
Released
-
date not published

Pricing

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

TierInput / 1MOutput / 1M
Standard$0.05$0.30
Cached input
50% off input
$0.025

What it costs a month

At ZeroGPU list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$35.25
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$28.80
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$57.00
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$192
Use your own numbers

Where to run it

1 host serve Llama 3.2 3B Instruct Fast. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
ZeroGPUCloud
llama-3.2-3b-instruct-fast
$0.05 / $0.30

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 ZeroGPU.

Smaller Meta models

No cheaper Meta model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • Nemotron 3.5 Lightning
    NVIDIA · on Nebius AI Studio
    $0.06 / $0.24
    Has every key capability
    7% less
  • MiniMax M3.1 Flash
    MiniMax · on Refuseless
    $0.05 / $0.25
    Has every key capability
    11% less
  • Qwen3.5 9B
    Alibaba Qwen · on EmpirioLabs AI
    $0.09 / $0.13
    Has every key capability
    11% less
  • Ministral 3B 2512
    Mistral AI · on Mistral AI
    $0.10 / $0.10
    Has every key capability
    11% less

Call it

Using ZeroGPU's API.

Python
from litellm import completion

response = completion(
    model="zerogpu/llama-3.2-3b-instruct-fast",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Llama 3.2 3B Instruct Fast cost?

On ZeroGPU, Llama 3.2 3B Instruct Fast costs $0.05 per million input tokens and $0.30 per million output tokens. Cached input costs $0.025 per million tokens.

Which APIs offer Llama 3.2 3B Instruct Fast?

Llama 3.2 3B Instruct Fast runs on 1 API: ZeroGPU.

What is Llama 3.2 3B Instruct Fast's context window?

131K tokens, with up to 8K tokens in one reply.

Is Llama 3.2 3B Instruct Fast being retired?

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

What are cheaper alternatives to Llama 3.2 3B Instruct Fast?

Models that cost less and keep most of what it can do: Nemotron 3.5 Lightning on Nebius AI Studio (7% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Llama 3.2 3B Instruct Fast's API model id?

Use llama-3.2-3b-instruct-fast with ZeroGPU.

Sources

Raw catalog entry for llama-3.2-3b-instruct-fast on ZeroGPU
{
  "cache_read_input_token_cost": 2.5e-8,
  "input_cost_per_token": 5e-8,
  "litellm_provider": "zerogpu",
  "max_input_tokens": 131072,
  "max_output_tokens": 8192,
  "mode": "chat",
  "output_cost_per_token": 3e-7,
  "source": "https://docs.zerogpu.ai/docs/model-catalog"
}