Last refreshed 2026-07-26. Next refresh: weekly.
Why use Claude 3 Haiku on Vercel AI Gateway?
Vercel AI Gateway offers Claude 3 Haiku with pay-as-you-go pricing at $0.25/1M input tokens. Vercel AI Gateway is a unified AI proxy providing a single OpenAI-compatible API endpoint to 275+ models from 25+ providers including Anthropic, OpenAI, Google, Meta, Mistral, DeepSeek, xAI, Alibaba, Amazon, ByteDance, Cohere, MiniMax, MoonshotAI, KwaiPilot, Black Forest Labs, Recraft, Voyage AI, NVIDIA, and more.
Compare Claude 3 Haiku across 7 providers to find the best fit for your use caseInput / 1M
$0.25
Output / 1M
$1.25
Cache
read $0.030
Batch
Not sourced
Setup recipe
Python + curlInstall
pip install openaiAuth
export AI_GATEWAY_API_KEY=...Call
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["AI_GATEWAY_API_KEY"],Model ID
anthropic/claude-3-haikuRequest example
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["AI_GATEWAY_API_KEY"],
base_url="https://ai-gateway.vercel.sh/v1"
)
response = client.chat.completions.create(
model="anthropic/claude-3-haiku",
messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)Gotchas
- Use provider model ID "anthropic/claude-3-haiku", not the LLMReference slug "claude-3-haiku".
- creator/model-name e.g. kwaipilot/kat-coder-pro-v2
- The examples expect AI_GATEWAY_API_KEY; rename it only if your application config maps the new variable.
Compare Claude 3 Haiku Across Providers
| Provider | Input (per 1M) | Output (per 1M) |
|---|---|---|
| AWS Bedrock | $0.25 | $1.25 |
| GCP Vertex AI | $0.25 | $1.25 |
| Salesforce Einstein Generative AI | — | — |
| Anthropic | $0.25 | $1.25 |
| OpenRouter | $0.25 | $1.25 |
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.25 |
| Output tokens | $1.25 |
Capabilities
VisionMultimodalReasoningStructured OutputsCode Execution
About Claude 3 Haiku
Claude 3 Haiku is Anthropic's Claude 3 model with multimodal text and image input and an optional reasoning mode. It is deprecated (originally released 2024-03-04); use it only for reproducing earlier results or evaluating drift over time.
Get Started
Model Specs
Released2024-03-04
Parameters20B
Context200k
ArchitectureDecoder Only
Knowledge cutoff2023-08