All tasks / Models
Claude Haiku 5.5
Anthropic's fastest, lowest-cost Claude tier for high-volume, latency-sensitive tasks.
- Anthropic
- Closed
- Released 7 Oct 2026
Where it ranks
- BChat & assistantsModels · Cloud · 43 on Artificial Analysis Intelligence Index (secondary benchmark)
- –Documents & OCRModels · Cloud · no result on this task’s benchmarks yet
- CCodingModels · Cloud · 1,586 on LMArena WebDev / Code Arena (overall)
- –Browser & computer agentsModels · Cloud · no result on this task’s benchmarks yet
- CWorkflow automationModels · Cloud · 35.4% on AutomationBench-AA (Artificial Analysis)
Facts
- Vendor
- Anthropic
- Family
- Claude Haiku
- Released
- 7 Oct 2026
- Weights
- Closed: API and apps only
- License
- proprietary
- API price
- $0.1 in / $0.5 out per 1M tokens
- Access
- Paid only
- Last verified
- 18 Oct 2026
Price from platform.claude.com, 18 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Artificial Analysis Intelligence Index | 43 | Claude Haiku 5.5 (Max)Raw value from the page's embedded model data; the site displays it rounded to an integer. | 18 Oct 2026 | artificialanalysis.ai |
| AutomationBench-AA (Artificial Analysis) | 35.4% | Claude Haiku 5.5 (Max)AutomationBench-AA Score (share of each task's objectives completed, a task with any guardrail violation scoring zero); raw value from the page's embedded model data (displayed rounded); rank 30 of 32; Tasks Completed 13.85%, Objectives Completed 75.64%; no last-updated date shown, so as_of = retrieval date. | 19 Oct 2026 | artificialanalysis.ai |
| LMArena WebDev / Code Arena (overall) | 1,586 | Claude Haiku 5.5 (High)From LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 32; 3376 votes; 95% CI 1574.9-1597.1. | 18 Oct 2026 | huggingface.co |
| ParseBench (models) | 57.4% | Anthropic Haiku 5.5Overall (mean of tables, charts, content faithfulness, semantic formatting, visual grounding) from leaderboard.csv in the ParseBench repo (last leaderboard commit 2026-10-09). | 9 Oct 2026 | github.com |
| SWE-Bench Pro V2 (public set, full) | 87.2% | Haiku 5.5 (mini-swe-agent) highSWE-Bench Pro V2 Full (642 tasks); row added 2026-10-08; 95% CI +/-2.36; harness in the variant name; read from the leaderboard page's embedded data. | 19 Oct 2026 | labs.scale.com |
Where you can use it
- SAmazon BedrockRuntimes & gateways · Cloud
- SBase44App builders · Cloud
- SBlinkApp builders · Cloud
- SClaudeApps · Cloud
- SClaude CodeCoding agents · Cloud
- –CodeGPTEditors & extensions · Local
- SCursorEditors & extensions · Cloud
- SDevin (Devin Desktop, formerly Windsurf)Editors & extensions · Cloud
- CFreebuff (formerly Codebuff)Coding agents · Cloud
- SGemini Enterprise Agent Platform (formerly Vertex AI)Runtimes & gateways · Cloud
- SGitHub CopilotEditors & extensions · Cloud
- SLiteLLMRuntimes & gateways · Local
- SLobeHub (formerly LobeChat)Apps · Local
- BOpenCodeCoding agents · Local
- SOpenRouterRuntimes & gateways · Cloud
- SPoeApps · Cloud
- SPortkey AI GatewayRuntimes & gateways · Local
- SRequestyRuntimes & gateways · Cloud
- SVeniceApps · Cloud
- SVercel AI GatewayRuntimes & gateways · Cloud
- –ZedEditors & extensions · Local
Earlier versions
- Claude Haiku 4.5 · 15 Oct 2025
Sources
- https://platform.claude.com/docs/en/about-claude/models/overview
- https://platform.claude.com/docs/en/about-claude/model-deprecations
- https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset
- https://artificialanalysis.ai/leaderboards/models
- https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool
- https://github.com/run-llama/ParseBench
- https://labs.scale.com/leaderboard/swe_bench_pro_public_v2
- https://artificialanalysis.ai/evaluations/automationbench-aa