Overview

Identity

Canonical model: Yesidentity confidence: high
Official checkpoints: None recorded — closed weights
Artifacts: None recorded
Provider deployments: 1
API aliases: anthropic/claude-haiku-4.5claude-4-5-haikuclaude-4-5-haiku-reasoningclaude-haiku-4-5-20251001
Folded evaluation variants: 1

Openness

Closed / proprietary— weights not available; 7 dimensions unknown.

Weights are not available; the model is reachable only through an API or a product.

Weights
No
Inference code
—
Training code
—
Training data
—
Dataset
—
Commercial use
—
Redistribution
—
Derivatives
—

dimensions marked null are unknown, not false

Key facts

Release date: Source:Claude Developer Platform docsT1observed 13 h agohigh
Status: Source:Claude Developer Platform docsT1observed 13 h agohigh
Knowledge cutoff: Source:Claude Developer Platform docsT1observed 13 h agohigh
Retirement date: Source:Claude Developer Platform docsT1observed 13 h agohigh
Official page: platform.claude.com/docs/en/models/haiku-4-5/overview
Source:Claude Developer Platform docsT1observed 13 h agohigh
API model id: Source:Claude Developer Platform docsT1observed 13 h agohigh
Openrouter id: Source:OpenRouter public model & pricing listingT2observed 13 h agomedium

Architecture

Tokenizer: Source:OpenRouter public model & pricing listingT2observed 13 h agomedium
Training data cutoff: Source:Claude Developer Platform docsT1observed 13 h agohigh

Capabilities

Modalities

Modalities: textimage
Input: textimage
Output: text

Capabilities

Tool calling
Yes
Claude Developer Platform docs · T1
Structured output
Yes
OpenRouter public model & pricing listing · T2
Reasoning
Yes
OpenRouter public model & pricing listing · T2
Vision
Yes
Claude Developer Platform docs · T1
Audio
Unavailable
Fine-tuning available
Unavailable

Context window: Source:Claude Developer Platform docsT1observed 13 h agohigh
Max output: Source:Claude Developer Platform docsT1observed 13 h agohigh
Knowledge cutoff: Source:Claude Developer Platform docsT1observed 13 h agohigh
Tokenizer: Source:OpenRouter public model & pricing listingT2observed 13 h agomedium

Benchmarks17

Compare with another model →

Comparable same task and conditions · Partially comparable same task, conditions differ (effort, temperature, judge) · Not comparable different variant or metric

Benchmark results grouped by comparability group
Benchmark · group	Trust	Configuration	Results	vs leader	Evaluated	Source
Terminal-Benchagentic · accuracy · variant=v2.1 · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisvariantv2.1reasoningonconditions differ across rows → partially comparable	1	non-primary group	obs. 11 Sept 2026	artificialanalysis.aiT2
Terminal-Benchagentic · accuracy · variant=hard · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisvarianthard	2	−38.6 ptvs gpt-5.6-sol	obs. 11 Sept 2026	artificialanalysis.aiT2
Terminal-Benchagentic · accuracy · variant=v4.0 · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisvariantv4.0reasoningonconditions differ across rows → partially comparable	1	non-primary group	obs. 11 Sept 2026	artificialanalysis.aiT2
τ²-benchagentic · pass^1 · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable	2	−44.4 ptvs Z.ai GLM 5.2	obs. 11 Sept 2026	artificialanalysis.aiT2
SciCodecoding · accuracy · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable	1	−20.8 ptvs Claude Fable 5.1	obs. 11 Sept 2026	artificialanalysis.aiT2
Artificial Analysis Intelligence Indexcomposite · index	Independent	reasoningonversion4.3conditions differ across rows → partially comparable	2	−35.8vs Claude Fable 5.1	obs. 11 Sept 2026	artificialanalysis.aiT2
IFBenchinstruction-following · accuracy · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable	2	−29.0 ptvs Grok 4.3	obs. 11 Sept 2026	artificialanalysis.aiT2
Humanity's Last Examknowledge · accuracy · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable	2	−48.8 ptvs Claude Fable 5.1	obs. 11 Sept 2026	artificialanalysis.aiT2
MMMU-Promultimodal · accuracy · evaluator=Artificial Analysis	Independent	evaluatorArtificial Analysisreasoningonconditions differ across rows → partially comparable	2	−28.3 ptvs gpt-6-astra	obs. 11 Sept 2026	artificialanalysis.aiT2
GPQA Diamondreasoning · accuracy · variant=GPQA Diamond · evaluator=Artificial Analysis	Independent	evaluatorArtificial AnalysisvariantGPQA Diamondreasoningonconditions differ across rows → partially comparable	2	−29.1 ptvs gpt-6-astra	obs. 11 Sept 2026	artificialanalysis.aiT2

Current rows only, grouped by benchmark → canonical metric → comparability group (task configuration). Effort variants folded into this model appear as rows of the same group. 17 current rows in total. “vs leader” compares with the current leader of the benchmark's primary group only; other groups are not directly comparable. Comparability rules →

Providers & Pricing3

All offers in the price terminal →

Provider deployments, cheapest output first
Provider	Context	Cached in	Cache write	Native units	Status	Observed	Source
Anthropic APIcheapest outputanthropic/claude-haiku-4.5:batch	200Kout 64K	$0.05	$0.625	—	active	3 h agosince 11 Sept 2026	openrouter.aiT2
Anthropic API	—	$0.10	$1.25	cache_write_1h_per_mtok=2	active	13 h agosince 11 Sept 2026	platform.claude.comT1
Anthropic APIanthropic/claude-haiku-4.5	200Kout 64K	$0.10	$1.25	—	active	3 h agosince 11 Sept 2026	openrouter.aiT2

USD per 1M tokens as published by each provider; native units (per-request fees, flex/priority tiers) are kept verbatim. Rows are append-only — every price change is kept in the history below. Cost of a workload →

Price history

Step lines per provider; amber markers are recorded changes. Click a marker or a row for the evidence behind that price.

Output price · USD / 1M tokens 1 provider

Anthropic API

Anthropic API$5 → $2.511 Sept 2026
Anthropic APIfirst observed $511 Sept 2026

Input price · USD / 1M tokens 1 provider

Anthropic API

Anthropic API$1 → $0.5011 Sept 2026
Anthropic APIfirst observed $111 Sept 2026

Versions & Artifacts0

Version history

Context windowfirst observation only

11 Sept 2026current

Knowledge cutofffirst observation only

11 Sept 2026current

Max outputfirst observation only

11 Sept 2026current

Opennessfirst observation only

11 Sept 2026current

Statusfirst observation only

11 Sept 2026current

Each hop is a claim: click a value for its source, tier and observation time. Nothing is overwritten — a new observation closes the previous claim.

Artifacts 0

No artifact (checkpoint, quantisation, conversion or packaging) points to this model yet.

Timeline1

Full timeline →

12 Sept 2026
Property changedModelClaude Haiku 4.5Anthropic
Claude Haiku 4.5: reasoning changed from false to true
ReasoningNo→Yesopenrouter
openrouter.ai

Change history34

Temporal, append-only claims: a new observation closes the previous claim instead of overwriting it. Rewind the record with the as-of picker.

Viewing AI Atlas as of 12 Mar 2026 — attributes exactly as the atlas knew them on that day; later corrections are not shown.

Back to today →

Claude Haiku 4.5 was not yet in AI Atlas on 12 Mar 2026

First seen 11 Sept 2026. Nothing is inferred backwards: no attribute is shown for dates before the first observation.

34 claims · 32 properties2 conflicting

Familyfamily1

Claim history for Family
Value	Valid from → to	Status	Source	Confidence	Extractor
Claude	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Release daterelease_date1

Claim history for Release date
Value	Valid from → to	Status	Source	Confidence	Extractor
15 Oct 2025	11 Sept 2026 20:52 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Statusstatus1

Claim history for Status
Value	Valid from → to	Status	Source	Confidence	Extractor
active	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Opennessopenness1

Claim history for Openness
Value	Valid from → to	Status	Source	Confidence	Extractor
proprietary	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Context windowcontext_length1

Claim history for Context window
Value	Valid from → to	Status	Source	Confidence	Extractor
200K tokens	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Max outputmax_output_tokens1

Claim history for Max output
Value	Valid from → to	Status	Source	Confidence	Extractor
64K tokens	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Knowledge cutoffknowledge_cutoff1

Claim history for Knowledge cutoff
Value	Valid from → to	Status	Source	Confidence	Extractor
Feb 2025	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Training data cutofftraining_data_cutoff1

Claim history for Training data cutoff
Value	Valid from → to	Status	Source	Confidence	Extractor
Jul 2025	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Modalitiesmodalities2conflicting claims

Claim history for Modalities
Value	Valid from → to	Status	Source	Confidence	Extractor
image, text	11 Sept 2026 21:10 UTC → current	conflicting	OpenRouter public model & pricing listingT2	medium	deterministic
text, image	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	conflicted	deterministic

Input modalitiesmodalities_input2conflicting claims

Claim history for Input modalities
Value	Valid from → to	Status	Source	Confidence	Extractor
image, text	11 Sept 2026 21:10 UTC → current	conflicting	OpenRouter public model & pricing listingT2	medium	deterministic
text, image	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	conflicted	deterministic

Output modalitiesmodalities_output1

Claim history for Output modalities
Value	Valid from → to	Status	Source	Confidence	Extractor
text	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Tokenizertokenizer1

Claim history for Tokenizer
Value	Valid from → to	Status	Source	Confidence	Extractor
Claude	11 Sept 2026 21:10 UTC → current	current	OpenRouter public model & pricing listingT2	medium	deterministic

API model idapi_model_id1

Claim history for API model id
Value	Valid from → to	Status	Source	Confidence	Extractor
claude-haiku-4-5-20251001	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

API aliasapi_alias1

Claim history for API alias
Value	Valid from → to	Status	Source	Confidence	Extractor
claude-haiku-4-5	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Retirement dateretirement_date1

Claim history for Retirement date
Value	Valid from → to	Status	Source	Confidence	Extractor
15 Oct 2026	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Retirement tentativeretirement_tentative1

Claim history for Retirement tentative
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Official pageofficial_url1

Claim history for Official page
Value	Valid from → to	Status	Source	Confidence	Extractor
https://platform.claude.com/docs/en/models/haiku-4-5/overview	11 Sept 2026 20:52 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Bedrock model idbedrock_model_id1

Claim history for Bedrock model id
Value	Valid from → to	Status	Source	Confidence	Extractor
anthropic.claude-haiku-4-5	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Claude aws model idclaude_aws_model_id1

Claim history for Claude aws model id
Value	Valid from → to	Status	Source	Confidence	Extractor
claude-haiku-4-5	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Default effortdefault_effort1

Claim history for Default effort
Value	Valid from → to	Status	Source	Confidence	Extractor
Not supported	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Descriptiondescription1

Claim history for Description
Value	Valid from → to	Status	Source	Confidence	Extractor
The fastest model with near-frontier intelligence	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

File inputfile_input1

Claim history for File input
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:10 UTC → current	current	OpenRouter public model & pricing listingT2	medium	deterministic

Foundry model idfoundry_model_id1

Claim history for Foundry model id
Value	Valid from → to	Status	Source	Confidence	Extractor
claude-haiku-4-5	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Latency tierlatency_tier1

Claim history for Latency tier
Value	Valid from → to	Status	Source	Confidence	Extractor
Fastest	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Openrouter idopenrouter_id1

Claim history for Openrouter id
Value	Valid from → to	Status	Source	Confidence	Extractor
anthropic/claude-haiku-4.5	11 Sept 2026 21:10 UTC → current	current	OpenRouter public model & pricing listingT2	medium	deterministic

Reasoningreasoning1

Claim history for Reasoning
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:10 UTC → current	current	OpenRouter public model & pricing listingT2	medium	deterministic

Structured outputstructured_output1

Claim history for Structured output
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:10 UTC → current	current	OpenRouter public model & pricing listingT2	medium	deterministic

Supported parameterssupported_parameters1

Claim history for Supported parameters
Value	Valid from → to	Status	Source	Confidence	Extractor
include_reasoning, max_completion_tokens, max_tokens, reasoning, response_format, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p	11 Sept 2026 21:10 UTC → current	current	OpenRouter public model & pricing listingT2	medium	deterministic

Thinkingthinking1

Claim history for Thinking
Value	Valid from → to	Status	Source	Confidence	Extractor
Extended	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Tool callingtool_calling1

Claim history for Tool calling
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Vertex model idvertex_model_id1

Claim history for Vertex model id
Value	Valid from → to	Status	Source	Confidence	Extractor
claude-haiku-4-5@20251001	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Visionvision1

Claim history for Vision
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 20:51 UTC → current	current	Claude Developer Platform docsT1	high	deterministic

Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →

Provenance

Attributed facts

35

Source tiers

T1T226 / 9

Freshest observation

3 h ago

Conflicts

8 flagged

Source documents 5

Source documents
Source	Document	Type	Tier	Last observed	Snapshots
Claude Developer Platform docs	docs.claude.com/en/docs/about-claude/pricing.md	pricing	T1· Official	5 h ago	1
Claude Developer Platform docs	docs.claude.com/en/docs/about-claude/models/overview.md	model_docs	T1· Official	5 h ago	1
Claude Developer Platform docs	platform.claude.com/docs/en/models/haiku-4-5/overview.md	model_page	T1· Official	13 h ago	1
OpenRouter public model & pricing listing	openrouter.ai/api/v1/models	catalogue	T2· Quality secondary	3 h ago	7
Artificial Analysis	artificialanalysis.ai/leaderboards/models	leaderboard	T2· Quality secondary	11 h ago	1

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.

Data quality (79/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →