whatmodel.sh / Claude Haiku 5.5

Claude Haiku 5.5

+ contribute

Claude Haiku 5.5 is a closed-weights language model from Anthropic, released October 2026. Licence: proprietary.

Catalogue identity notes

Official Anthropic model claude-haiku-5-5 (Claude API alias = pinned dateless snapshot; Bedrock anthropic.claude-haiku-5-5), 'Released October 7, 2026', status 'Active (latest)'. Announced alongside Sonnet 5.5 on 2026-09-28. Text and images -> text, adaptive thinking (default effort medium). Scores below are from its system card at max effort. Not the earlier claude-haiku-4-5.

Model information sources

Release date ↗

77.4±0.7#22 of 719 LLMsfrom 15 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of Claude Haiku 5.5, out of 828 with a score, then more from Anthropic.

#	model	score	where	RAM	released
20	GPT-5.4 OpenAI	78±1	api $5.63/M		2026-03
21	DeepSeek V4.1 Flash DeepSeek	78±1	api $0.26/M +local	~537GB	2026-09
22	Claude Haiku 5.5 Anthropic	77±1	api $0.20/M		2026-10
23	Gemini 3.8 Flash Google	77±1	api $1.50/M		2026-09
24	Muse Spark 1.3 Meta	77±1	api $2.00/M		2026-09
	more from Anthropic
2	Claude Opus 5.5 Anthropic	87±1	api $8.00/M		2026-09
4	Claude Fable 5.1 Anthropic	85±1	api $20.00/M		2026-09
5	Claude Sonnet 5.5 Anthropic	84±1	api $4.00/M		2026-09

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #13 of 300 · 28% of its score · 5 tests

Beats GPT-5.5 Pro, GPT-5.4 Pro, GPT-5.5 here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Pro (split unspecified) lab	64.8% ✎	▄█·▄·▄██···▄ beats 2/9	very hard · avg 7.6%	▇▇▇▇ 10%	sharp · at its level
Terminal-Bench 4.0 lab	39.2% ✎	▄·····▄··███ beats 2/7	very hard · avg 0.5%	▇▇ 5.9%	sharp · at its level
SWE-bench Multilingual lab	83.7% ✎	▃··▃·▅▃▃█··▃ beats 2/9	hard · avg 31%	▇▇ 5.1%	typical
Chartography with tools lab	86.2% ✎	▃········▃██ beats 4/7	very hard · avg 4.7%	▇▇ 4.4%	sharp · at its level
Humanity's Last Exam with tools lab	57.4% ✎	▁······▁·▄·█ beats 3/10	very hard · avg 14%	▇ 2.7%	at its level

coding▁▁▁▃▆██▇▆▄▃▁ #30 of 458 · 23% of its score · 3 tests

Beats GPT-5.5, GPT-5.6 Terra, Claude Opus 4.8 here, though they rank higher overall. Trails GLM-5.2, GLM-5.3-Flash, Grok 4.7 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Pro (split unspecified) lab	64.8% ✎	▄█·▄·▄██···▄ beats 2/9	very hard · avg 7.6%	▇▇▇▇ 10%	sharp · at its level
Arena WebDev (Code Arena)	1586.9 ✎	▁▂▄▃▅█▄▇▆▄▁▁ beats 93/119	avg 981	▇▇▇ 7.6%	sharp
SWE-bench Multilingual lab	83.7% ✎	▃··▃·▅▃▃█··▃ beats 2/9	hard · avg 31%	▇▇ 5.1%	typical

math▁▂▃▅▆▅▄█▄▃▁▁ #20 of 398 · 37% of its score · 3 tests

Beats DeepSeek V4.1 Flash here, though they rank higher overall. Trails GPT-5.6 Luna, GPT-6 Luna here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
FrontierMath tier 4 v2	46.3% ✎	▁▂▁▂█▅▂▂▁▁▁▁ beats 44/63	very hard · avg 0.1%	▇▇▇▇▇▇▇▇ 20%	sharp · at its level
FrontierMath tiers 1–3 v2	75.1% ✎	▂··▁▁▃▄▄█▅▃▃ beats 65/82	very hard · avg 1.2%	▇▇▇▇▇▇▇ 17%	sharp · at its level
OTIS mock AIME 2024–2025	98.9% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 177/193	very hard · avg 8.5%	▇ 0.5%	sharp · maxed out · noisy

web search▂▁▂▆▄█▃▃▁▁·▁ #28 of 80 * · 2.7% of its score · 1 test

Trails GPT-4o mini (2024-07-18), Mistral Small 3.1 24B (2503), GPT-4o (2024-08-06) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Humanity's Last Exam with tools lab	57.4% ✎	▁······▁·▄·█ beats 3/10	very hard · avg 14%	▇ 2.7%	at its level

other tests6 not in a task · 27% of its score

test	result	vs the others in its run	difficulty	counts	why
Furniture Assembly (Epoch)	47.5% ✎	█▆▆▆▆▃▅▂▂▂·▅ beats 21/30	very hard · avg 1.2%	▇▇▇ 7.7%	sharp · at its level
Mystery Game Puzzles (Epoch)	30.0% ✎	█··▂▃▇▄█▂▃▁▁ beats 48/74	very hard · avg 9.4%	▇▇▇ 6.9%	sharp · at its level · noisy
Humanity's Last Exam lab	45.9% ✎	▂·····▂·▄▂█▄ beats 3/10	very hard · avg 5.8%	▇▇ 5.3%	at its level
GPQA Diamond	89.6% ✎	▁▁▁▁▂▄▆▄▅▇█▅ beats 169/208	hard · avg 40%	▇ 2.5%	sharp
Global-MMLU lab	87.8% ✎	▄▄····▄▄··▄█ beats 0/6	medium · avg 63%	▇ 2.5%	typical
SimpleQA Verified	23.8% ✎	▁▁▃▂▃▃▆▆█▃▂▅ beats 15/78	very hard · avg 8.0%	▇ 1.7%	at its level

$ where to run it

Through an API

api · $0.20/M · $0.1 in / $0.5 out per 1M tokens · 1M context · platform.claude.com ↗ · price sourceObserved 2026-10-09 on Anthropic's own Claude Haiku 5.5 model page (Pricing table; same rows on platform.claude.com/docs/en/about-claude/pricing): USD per million tokens, standard uncached input/output 'for prompts up to 100,000 tokens'. Prompts over 100,000 tokens cost $0.50 input / $2.50 output; batch (50% off) and cache rates excluded.

$ how the score was worked out · Epoch AI — internal evaluatio… 56%, Anthropic system cards and mo… 36%, LMArena leaderboard dataset 8%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

source	results	share of the score	licence
Epoch AI — internal evaluations	7	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 56%	CC BY 4.0
Anthropic system cards and model announcements (lab-reported scores) · lab-reported	7	▇▇▇▇▇▇▇▇▇▇▇ 36%	Factual vendor-reported score (not licensed content)
LMArena leaderboard dataset	1	▇▇ 8%	CC BY 4.0
prior · pull toward the middle		▇ 0%
weighted average		= 77.4