Claude Haiku 4.5
Claude Haiku 4.5 is the live cheap / fast Claude row on the models overview, not leftover Claude 3 Haiku. Rechecked 2026-08-18: the Claude models overview prints Haiku 4.5 at $1 / $5 per 1M tokens and 200k context, beside Sonnet 5 ($2 / $10, 1M), Opus 5 ($5 / $25, 1M), and Fable 5 ($10 / $50, 1M). Do not call leftover 3 Haiku the current fast default.
Compare Claude Sonnet 5 if you wanted the live mid-tier 1M row, Claude Opus 5 if you wanted the live Opus path, and leftover Claude 3 Haiku if you still had the old fast bookmark.
Key Features
- Live cheap Claude row: First-party table prints Haiku 4.5 at $1 / $5 and 200k context.
- Not leftover 3 Haiku: The older leftover page stays a previous snapshot.
- Not the 1M default: 1M context on the same table sits on Sonnet 5 / Opus 5 / Fable 5.
- Website is the models overview: Recheck that table before you hard-code an id.
Use Cases
- People who need the current cheap Claude row after leftover 3 Haiku was rewritten.
- People comparing Haiku 4.5 vs Sonnet 5 vs leftover 3 Haiku.
- People who were about to paste 3 Haiku as Anthropic's current fastest model.
Limitation: We did not invent a later Haiku slug or a 2026 bench crown. Recheck the overview for the exact API id.
Pricing
First-party pages on 2026-08-18.
| Piece | Price | Notes |
|---|---|---|
| Claude Haiku 4.5 | $1 / $5 per 1M | 200k context on the models overview. |
| Claude Sonnet 5 | $2 / $10 per 1M | 1M context. Claude Sonnet 5. |
| Leftover 3 Haiku | Previous snapshot | Claude 3 Haiku. |
If a leftover blog still calls 3 Haiku the current fast Claude, treat this page as the correction.
Getting Started
- Open the models overview and confirm the Haiku 4.5 row.
- Recheck Claude Sonnet 5 if you needed 1M context.
- Recheck leftover Claude 3 Haiku only as a historical bookmark.
- Do not paste leftover 3 Haiku latency copy onto Haiku 4.5.
Frequently Asked Questions
Is this the same as Claude 3 Haiku?
No. Claude 3 Haiku is a leftover previous snapshot.
Is Haiku 4.5 the 1M Claude default?
No. The 1M rows on the same table are Sonnet 5, Opus 5, and Fable 5.
Same as Claude Sonnet 5?
No. Claude Sonnet 5 is the live mid-tier $2 / $10 row.
Alternatives
- Claude Sonnet 5: Live mid-tier 1M Claude row.
- Claude Opus 5: Live Opus path at $5 / $25.
- GPT-5.6 Luna: Live cheap OpenAI sibling if you wandered here for a current low-cost model.
Tips
- Call Haiku 4.5 the live cheap / fast Claude row, not leftover 3 Haiku.
- Recheck the overview for the exact API id before you pin production.
- Do not invent a Haiku 4.5 bench crown.
Conclusion
Claude Haiku 4.5 is the live cheap / fast Claude row at $1 / $5 and 200k context. Start at the models overview, then decide whether Claude Sonnet 5 already covers a longer context path.
Comments
No comments yet. Be the first to comment!
Related Tools
Claude Sonnet 5
docs.claude.com/en/docs/about-claude/models/overview
Claude Sonnet 5: live mid-tier Claude row at $2/$10 per 1M tokens and 1M context. Rechecked 2026-08-18. Not leftover Sonnet 4.5.
Claude v1
docs.claude.com/en/docs/about-claude/models/overview
Previous Claude v1 snapshot. Rechecked: live docs lead with Fable 5, Opus 5, Sonnet 5, and Haiku 4.5. Not a 2026 text default.
Claude 3 Haiku
docs.claude.com/en/docs/about-claude/models/overview
Previous Claude 3 Haiku snapshot. Rechecked: live docs lead with Claude 5 / 4.6. Not the 2026 fast default.
Related Insights
Anthropic Subagent: The Multi-Agent Architecture Revolution
Deep dive into Anthropic multi-agent architecture design. Learn how Subagents break through context window limitations, achieve 90% performance improvements, and real-world applications in Claude Code.
Complete Guide to Claude Skills - 10 Essential Skills Explained
Deep dive into Claude Skills extension mechanism, detailed introduction to ten core skills and Obsidian integration to help you build an efficient AI workflow
The Twilight of Low-Code Platforms: Why Claude Agent SDK Will Make Dify History
A deep dive from first principles of large language models on why Claude Agent SDK will replace Dify. Exploring why describing processes in natural language is more aligned with human primitive behavior patterns, and why this is the inevitable choice in the AI era.