Claude Haiku 4.5
Fast Tier Reasoning Capable
Anthropic's October 2025 small/fast tier release. The notable capability: Haiku 4.5 has extended-thinking mode at near-Sonnet quality on math and code benchmarks, while maintaining Haiku-tier pricing and latency. Effectively the workhorse for high-volume agentic workloads where Sonnet pricing was prohibitive.
Intelligence
Medium
Speed
Medium
128 tok/s output
Cost
High
$1.25 in / $5.00 out
Context
200K
Up to 200,000 tokens
How are Intelligence, Speed & Cost bucketed?
Intelligence and Speed buckets are percentile ranks on
Artificial Analysis. Cost buckets are fixed dollar
thresholds keyed off output-token price ($/M out).
Intelligence
- Top 1%≤ 1%
- Top 5%≤ 5%
- Top 10%≤ 10%
- Good≤ 25%
- Medium≤ 50%
- Below avg> 50%
Speed
- Top 1%≥ 345 tok/s
- Top 5%≥ 237 tok/s
- Top 10%≥ 196 tok/s
- Good≥ 146 tok/s
- Medium≥ 90 tok/s
- Slow< 90 tok/s
Cost
- Freeopen weights · self-host
- Low< $1 / M out
- Moderate$1–5 / M out
- High≥ $5 / M out
Why it matters
Haiku 4.5 is the model that makes "AI agents running 24/7" economically viable for most use cases. Where Opus is the capability ceiling and Sonnet is the production workhorse, Haiku 4.5 is the "let it run all the time" tier.
Core Capabilities
Long Documents
Handles entire codebases, books, and multi-doc RAG.
Multimodal
Combines text, vision, and audio in one model.
Generative
Produces images, video, audio, or other media.
Agent Workflows
Built for tool use and autonomous tasks.
Context Window
200k tokens
≈ 154+ pages
4k Chat 聊天
32k Long docs 长文档
128k Books 整本书
400k Multi-doc 多文档
1M Codebase 整个代码库
10M
200k
Availability
API
Available
Product / App
Available
Open Source
Not released
Enterprise
Contact sales
Pricing Model
Pay per token
Input and output billed separately.
Pay-per-token Capability / Performance
Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).
Lower 20% Upper 80% This model
Context / memory
Context window size · log-scaled
6.0
9.0
6.7
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this
Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.
What it feels like
- 73.3% on SWE-bench Verified — almost matches Sonnet 4 (state-of-the-art six months earlier) at 1/3 the price
- Responses that took 3-5s on Sonnet come back in <2s on Haiku — dramatically faster real-world feel
- $1/M input, $5/M output — 3x cheaper than Sonnet 4.5; sub-cent answers for typical chat turns
- First Haiku with extended thinking, computer use, and context awareness baked in
- Caylent: 'covers 70-90% of daily workloads at far better speed and cost' — replacement, not just supplement
- 200K context + 64K output — generous limits for the price tier
Best use cases
- High-volume customer-facing chatbots and support tooling
- Multi-agent swarms where many parallel small tasks beat one big call
- Background coding tasks (linting, fixups, docstring writing) that don't need Sonnet/Opus
- Mobile / edge-adjacent inference where every token cost compounds
Tools to try
Not ideal for
- The hardest reasoning, multi-system bug fixes, or long agent loops (use Sonnet 4.5+ or Opus)
- Frontier benchmark leaderboards — Haiku is intentionally one tier down
- Workloads needing tools/extended thinking on every call (cost flips vs Sonnet)