Bonus: Top up now and we'll double your first deposit — get x2 credits instantly.
Get Started
Sign Up
Sign Up

Anthropic: Claude Haiku 4.5 API: Endpoint & 30% Discount

Claude Haiku 4.5 is Anthropic’s fastest, most cost-efficient Claude 4.5-generation model, offering near-frontier intelligence with low latency and pricing optimized for large-scale use.

Browse available models

What is Claude Haiku 4.5?

Claude Haiku 4.5 is a small, fast large language model from Anthropic’s Claude 4.5 family, optimized for low-latency, cost-efficient deployment. It is mainly used for real-time conversational agents, support-style chatbots, and interactive applications that require quick responses at scale, as well as for production workloads like large-scale financial analysis and research where throughput and price are critical. It is also widely used for software engineering workflows, including code generation, debugging, and multi-agent coding or computer-use tasks, aided by its 200k-token context window, tool use, and vision support. Claude Haiku 4.5 belongs to the Claude 4.5 model family and succeeds earlier small models such as Claude 3.5 Haiku.

Can you self-host Claude Haiku 4.5?

No. We found no public weights for Claude Haiku 4.5 on Hugging Face, so there is nothing to download, quantize or serve on your own GPUs: no VRAM budget, no vLLM or GGUF build to plan for. The only way to run it is over an API.

Not on LLM.API yet

We do not route this model through our API at the moment. Browse the models you can call today — most workloads have a close match already live.

Browse available models

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Exacto (highest tool-calling accuracy).

List price by provider

InputOutput
Azure$0.001 in
$0.005 out
Amazon Bedrock$0.001 in
$0.005 out
Google$0.001 in
$0.005 out
Anthropic$0.001 in
$0.005 out
Amazon Bedrock$0.0011 in
$0.0055 out
Google$0.0011 in
$0.0055 out
Amazon Bedrock$0.0011 in
$0.0055 out
Google$0.0011 in
$0.0055 out

Provider list prices; the LLM.API discount applies on top.

Provider Input Output Cache read /M Latency Throughput Uptime
Azure30% offin $0.001; out $0.005 per 1M tokens200K tokensvision, tools, streaming, reasoning, JSON, structured
Amazon Bedrock30% offin $0.001; out $0.005 per 1M tokens200K tokensvision, tools, streaming, reasoning, JSON, structured
Google30% offin $0.001; out $0.005 per 1M tokens200K tokensvision, tools, streaming, reasoning, structured
Anthropic30% offin $0.001; out $0.005 per 1M tokens200K tokensvision, tools, streaming, reasoning, JSON, structured
Amazon Bedrock30% offin $0.0011; out $0.0055 per 1M tokens200K tokensvision, tools, streaming, reasoning, JSON, structured
Google30% offin $0.0011; out $0.0055 per 1M tokens200K tokensvision, tools, streaming, reasoning, structured
Amazon Bedrock30% offin $0.0011; out $0.0055 per 1M tokens200K tokensvision, tools, streaming, reasoning, JSON, structured
Google30% offin $0.0011; out $0.0055 per 1M tokens200K tokensvision, tools, streaming, reasoning, structured

Prices, context and availability from the OpenRouter public catalogue (this model is not served through LLM.API), updated nightly. Last updated 24 Sept 2026.

Try this model

Test Claude Haiku 4.5 right here — free to start.

Claude Haiku 4.5
Hi! Want to test the model?

Suggestions for your first prompt

Not available on LLM API yet

We do not route this model through our API at the moment, so there is no endpoint or code snippet for it yet. Browse the models you can call today — most workloads have a close match already live.

Browse available models

Uptime

30-Day Uptime
97.55%
Past Incidents (30d)
13
Error rate (24h)
8.20%

Last 30 days

19/30 days operational | 97.55% uptime

30 days ago Today
Operational Degraded Outage Maintenance
See All Incidents

Anthropic: Claude Haiku 4.5 in production

What the model is good at.

  • Conversational Chat

    Handles multi-turn conversations, follows instructions, answers questions, and maintains context for helpful, concise assistant-style dialogue.

  • Code Reasoning

    Understands and writes code, explains programming concepts, and assists with debugging and small-scale software or script tasks.

  • Image Understanding

    Interprets images, identifying objects, text, layouts, and visual relationships to support analysis and question answering.

  • Visual Text Extraction

    Reads and extracts text from images, screenshots, and scanned documents for downstream processing, search, or transformation.

  • Language Translation

    Translates between multiple natural languages while preserving meaning and tone across general-purpose text content.

What you get when you call Anthropic: Claude Haiku 4.5 through LLM.API.

  • Unified AI Routing

    Dynamically route each request to the optimal model across providers based on latency, cost, and quality. One API, pluggable policies, zero vendor lock‑in.

    One endpoint, any model

  • Cost-Aware Orchestration

    Set per-request or per-project cost policies and let LLM.API choose cheaper equivalents automatically. Eliminate manual price tuning while keeping predictable spend.

    Control spend by design

  • Automatic Fallbacks

    Configure multi-provider failover so requests seamlessly retry on backup models when a vendor is down or throttled. Ship resilient AI features without custom glue code.

    Resilience by default

  • Full-Stack Observability

    Get centralized traces, logs, metrics, and cost breakdowns across all models and vendors. Debug prompts, spot regressions, and optimize performance from a single dashboard.

    See every token

  • Task-Level Abstractions

    Describe tasks—chat, RAG, tool use, scoring—once and let LLM.API pick the right models and parameters. Iterate on behavior, not low-level API wiring.

    Code tasks, not glue

  • High-Throughput Batch

    Submit massive batches of requests through a unified endpoint with queueing, parallelism, and retries handled for you. Maximize throughput while staying within provider limits.

    Millions of calls, one API

When to Use — When NOT to Use

Use it if...

  • You need a low-cost, fast model for everyday assistant-style questions and answers.
  • You need to power high-volume chatbots where latency and affordability matter more than depth.
  • Your use case involves lightweight content generation like short emails, social posts, or summaries.
  • Your use case involves simple code edits, small bug fixes, or clarifying existing snippets.
  • You need a safe, guarded model for end-user applications with strong default alignment.
  • Your use case involves classification, tagging, or routing tasks on large text datasets.
  • You need a general-purpose helper embedded in products where cost must stay predictable.

Avoid if...

  • You need state-of-the-art reasoning on complex multi-step problems or intricate planning tasks.
  • Your workload requires advanced code generation for large projects or complex software architectures.
  • You need top-tier performance on long-context analysis like large research paper sets.
  • Your workload requires creative writing at flagship quality, such as novels or screenplays.
  • You need cutting-edge multimodal reasoning or image understanding beyond basic or experimental capabilities.
  • Your workload requires best-available performance on math-heavy, symbolic, or formal reasoning benchmarks.
  • You need highly specialized domain expertise comparable to premium, large-scale flagship language models.

COMMUNITY

What developers say about Claude Haiku 4.5

LMArena#151 · Elo 1396Arena coding#87 · Elo 1454Speed—Sources: LMArena, Artificial Analysis

Summarised from public developer discussions and write-ups. Opinions are the sources’, not LLM.API’s.

  • Developers on Hacker News report that the model is notably precise in code changes, avoiding irrelevant code sections more effectively than some competing frontier models during preliminary testing.
  • Engineers describe the model as a highly efficient 'scout' for mapping codebases and fast repository analysis, though some caution against using it for unsupervised, complex autonomous building tasks.
  • Developer reviews on Medium and Reddit highlight its significant speed and cost-efficiency, noting it can deliver coding performance similar to previous frontier tiers at a fraction of the price.

SOURCES

Frequently Asked Questions

  • What is Claude Haiku 4.5?

    Claude Haiku 4.5 is Anthropic’s fast, lightweight Claude 4.5-series model optimized for low-latency, low-cost text and vision use cases.

  • What is Claude Haiku 4.5 best suited for?

    Claude Haiku 4.5 is best for high-volume workloads like chatbots, data processing, RAG, small agents, and rapid vision tasks where speed and price matter.

  • What context window does Claude Haiku 4.5 support via LLM.API?

    Via LLM.API, Claude Haiku 4.5 supports up to a 200K token context window for input and conversation history.

  • How fast is Claude Haiku 4.5 on LLM.API?

    Claude Haiku 4.5 is designed for very low latency, typically returning first tokens in well under a second for short prompts.

  • What modalities does Claude Haiku 4.5 support?

    Claude Haiku 4.5 supports text input and output plus image input, enabling multimodal reasoning over documents, screenshots, and photos.

  • How is Claude Haiku 4.5 priced when used through LLM.API?

    Claude Haiku 4.5 is exposed through LLM.API’s own metered pricing, which may differ from Anthropic’s direct per-token rates.

  • How do I call Claude Haiku 4.5 through the LLM.API gateway?

    You select the Claude Haiku 4.5 model identifier in LLM.API requests, send prompts using the unified schema, and receive responses in a standard format.

  • How does Claude Haiku 4.5 compare to Claude Sonnet 4.5?

    Compared to Claude Sonnet 4.5, Haiku 4.5 is cheaper and faster but somewhat weaker on complex reasoning and highly advanced tasks.

  • What are the main limitations of Claude Haiku 4.5?

    Claude Haiku 4.5 can hallucinate, struggle with very complex reasoning, and should not be solely trusted for safety-critical or legally binding decisions.

  • Can Claude Haiku 4.5 handle streaming responses on LLM.API?

    Yes, Claude Haiku 4.5 supports token streaming through LLM.API so you can start processing output before the full response is generated.

COMPARE

Competitive Models

Claude Opus 4.7 is Anthropic’s most capable generally available large language model, designed for advanced coding, long-horizon agentic workflows, and high-resolution vision tasks. It emphasizes stronger multi-step reasoning, reliability on complex work, and improved instruction following compared to earlier Opus releases.

  • Claude Opus 4.8

    Up to 30%
    • Instruction Following

    Claude Opus 4.8 is a large language model from Anthropic’s Claude family, designed for high-level reasoning, detailed writing assistance, and complex problem solving. It emphasizes helpfulness, safety, and reliability across a wide range of professional and creative tasks.

  • Claude Sonnet 4.5

    Up to 30%
    • Text Generation

    Claude Sonnet 4.5 is an Anthropic large language model optimized for software development, computer use, and agentic workflows, offering strong performance on coding and reasoning tasks at mid-tier pricing. It is part of the Claude 4.5 generation and is available through multiple cloud providers and enterprise platforms.

    • Get one key to every model

    Swap your API key. Keep your code.

    Browse available models Talk to sales