---
format: "aidr-story-markdown/v1"
id: "b707014dc48224fdab2a810470ef1a26924c2ac150ca4f10cbcb64854b462413"
canonical_url: "https://aidr.today/b707014d?lang=en"
title: "Anthropic Haiku 5.5 Beats GPT-6 Luna on Every Reported Benchmark"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-07T19:39:45.000Z"
category: "Tools"
topics: ["anthropic","llm","devtools"]
source_urls: ["https://huggingnews.com/ai/anthropic-haiku-55-beats-gpt-6-luna-on-every-reported-benchmark-61342a25","https://x.com/claudeai/status/2107894039626277339","https://x.com/mark_k/status/2107904785286615399","https://x.com/rohanpaul_ai/status/2107897102886150300","https://x.com/ArtificialAnlys/status/2107911905822351609","https://x.com/kimmonismus/status/2107890621834990038","https://x.com/ns123abc/status/2107904361313517624","https://x.com/theo/status/2107914587354063142"]
summary: "Anthropic's Claude Haiku 5.5 outperformed OpenAI's GPT-6 Luna on every reported benchmark, including the OSWorld 2.1 computer use test, agentic coding, and visual reasoning. The model features an expanded 1M token context window and tiered pricing that costs $0.10 per 1M input tokens and $0.50 per 1M output tokens for prompts under 100k, which is 90% cheaper than Haiku 4.5. Rates increase fivefold to $0.50 and $2.50 for requests exceeding 100k tokens, and the model uses approximately 3x more output tokens than GPT-6 Luna to reach similar intelligence index scores."
---

# Anthropic Haiku 5\.5 Beats GPT\-6 Luna on Every Reported Benchmark

> [Open the canonical story](<https://aidr.today/b707014d?lang=en>)

**Published:** 2026-10-07T19:39:45.000Z
**Category:** Tools
**Topics:** anthropic, llm, devtools

## Summary

Anthropic's Claude Haiku 5\.5 outperformed OpenAI's GPT\-6 Luna on every reported benchmark, including the OSWorld 2\.1 computer use test, agentic coding, and visual reasoning\. The model features an expanded 1M token context window and tiered pricing that costs $0\.10 per 1M input tokens and $0\.50 per 1M output tokens for prompts under 100k, which is 90% cheaper than Haiku 4\.5\. Rates increase fivefold to $0\.50 and $2\.50 for requests exceeding 100k tokens, and the model uses approximately 3x more output tokens than GPT\-6 Luna to reach similar intelligence index scores\.

## Sources

- [Story source](<https://huggingnews.com/ai/anthropic-haiku-55-beats-gpt-6-luna-on-every-reported-benchmark-61342a25>)
- [Story source](<https://x.com/claudeai/status/2107894039626277339>)
- [Supporting source](<https://x.com/mark_k/status/2107904785286615399>)
- [Supporting source](<https://x.com/rohanpaul_ai/status/2107897102886150300>)
- [Supporting source](<https://x.com/ArtificialAnlys/status/2107911905822351609>)
- [Supporting source](<https://x.com/kimmonismus/status/2107890621834990038>)
- [Supporting source](<https://x.com/ns123abc/status/2107904361313517624>)
- [Supporting source](<https://x.com/theo/status/2107914587354063142>)