impulse

Claude 3 Haiku

Frontier lab
API id claude-3-haikuReleased 7 Mar 2024 (Mar 2024)
Input / 1M
$0.25
at Vertex AI
Output / 1M
$1.25
at Vertex AI
Context window
200K
tokens
Max output
4K
tokens per reply
Hosts
3
5 listings
Released
Mar 2024
7 Mar 2024

Pricing

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

TierInput / 1MOutput / 1M
Standard$0.25$1.25

What it costs a month

At Vertex AI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$169
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$165
  • Coding agent1,000 requests a day, 40K in / 2K out$375
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$900
Use your own numbers

Where to run it

3 hosts serve Claude 3 Haiku. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
claude-3-haiku
$0.25 / $1.25
claude-3-haiku
$0.25 / $1.25
us-gov-east-1 · 2024-03-07
anthropic.claude-3-haiku-20240307-v1:0
$0.30 / $1.50
+20% vs headline
Show 2 more listings: regions, snapshots and variants
Vertex AI
2024-03-07
claude-3-haiku@20240307
$0.25 / $1.25
Amazon Bedrock
us-gov-west-1 · 2024-03-07
anthropic.claude-3-haiku-20240307-v1:0
$0.30 / $1.50
+20% 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
TextImages
Produces
Text
Features
Assistant prefill

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 Vertex AI.

Smaller Anthropic models

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    60% less

Other frontier labs

  • GPT-5.4 Nano
    OpenAI · on OpenAI
    $0.20 / $1.25
    Has every key capability
    7% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    60% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    48% less

Open-weight picks

  • Qwen3.7 Plus
    Alibaba Qwen · on Replicate
    $0.276 / $1.10
    Has every key capability
    4% less
  • Inkling 2 Small
    Independent · on Cloudflare Workers AI
    $0.30 / $1.00
    Has every key capability
    5% less
  • GLM 4 6v
    Zhipu AI · on Novita AI
    $0.30 / $0.90
    Has every key capability
    10% less
  • MiMo V2.6 Pro
    Xiaomi · on GMI Cloud
    $0.348 / $0.696
    Has every key capability
    13% less

Model line

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

  1. Oct 2026Claude Haiku 5.5Latest$0.10 / $0.50
  2. Oct 2025Claude Haiku 4.5$1.00 / $5.00
  3. Oct 2024Claude 3.5 Haiku$0.80 / $4.00
  4. Mar 2024Claude 3 HaikuThis page$0.25 / $1.25

Call it

Using Vertex AI's API.

Python
from litellm import completion

response = completion(
    model="vertex_ai/claude-3-haiku",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Claude 3 Haiku cost?

On Vertex AI, Claude 3 Haiku costs $0.25 per million input tokens and $1.25 per million output tokens.

Which APIs offer Claude 3 Haiku?

Claude 3 Haiku runs on 3 APIs: Vertex AI, Vercel AI Gateway and Amazon Bedrock.

What is Claude 3 Haiku's context window?

200K tokens, with up to 4K tokens in one reply.

Is Claude 3 Haiku being retired?

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

Is there a newer version of Claude 3 Haiku?

Yes. Claude Haiku 5.5 is the newest model in this line, released on 7 Oct 2026.

What are cheaper alternatives to Claude 3 Haiku?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (60% less), GPT-5.4 Nano on OpenAI (7% less), GPT-6 Luna on OpenAI (60% less) and Qwen3.7 Plus on Replicate (4% less). Savings are on a mix of 3 input tokens to 1 output token.

When was Claude 3 Haiku released?

Anthropic released it on 7 Mar 2024.

What is Claude 3 Haiku's API model id?

Use claude-3-haiku with Vertex AI. Other hosts use their own ids: anthropic.claude-3-haiku-20240307-v1:0 on Amazon Bedrock.

Sources

Raw catalog entry for claude-3-haiku on Vertex AI
{
  "deprecation_date": "2026-08-23",
  "input_cost_per_token": 2.5e-7,
  "litellm_provider": "vertex_ai-anthropic_models",
  "max_input_tokens": 200000,
  "max_output_tokens": 4096,
  "mode": "chat",
  "output_cost_per_token": 0.00000125,
  "supports_assistant_prefill": true,
  "supports_function_calling": true,
  "supports_tool_choice": true,
  "supports_vision": true
}