How does Claude 3 Haiku rank among AI models?
Claude 3 Haiku is a superseded version and is not on the current board as of 22 September 2026. Its archive position, where one exists, is shown on the page; the newer version carries the current rank.
Anthropic·anthropic/claude-3-haiku
Claude 3 Haiku is ModelCap archive rank #315 (Measured), from the ModelCap Index as of 22 Sept 2026, 02:49 UTC.
Claude 3 Haiku listed OpenRouter input price is $0.25 per 1M tokens and listed output price is $1.25 per 1M tokens, from OpenRouter listed prices as of 22 Sept 2026, 02:49 UTC.
Claude 3 Haiku has a published context limit of 200K tokens in the ModelCap catalogue as of 22 Sept 2026, 02:49 UTC.
Claude 3 Haiku has 1 published OpenRouter serving endpoint, from OpenRouter as of 22 Sept 2026, 02:49 UTC.
Claude 3 Haiku weight access is api only, from ModelCap classification of published repository metadata as of 22 Sept 2026, 02:49 UTC.
Snapshot facts · download the public dataset · methodology
ModelCap Index · All-versions archive
15.1#315
Measured · 63% support · score interval 10.1–20.1
Evidence as of 13 Sept 2026, 00:00 UTC
Superseded by Claude Haiku 4.5. It no longer holds a current-board position; its all-versions archive rank is #315.
Input
$0.25
per 1M tokens
Output
$1.25
per 1M tokens
Context
200K
tokens
Providers
1
endpoint
Gateway spend
—
latest · Vercel share
Providers
Uptime measured over the last 30 minutesOther versions (1)
ModelCap Index · All-versions archive
Methodology15.1
Measured · 2 public benchmark observations across 2 boards
All-versions archive
#315
- Public rank basis
- Measured (measured)
- Index support
- 62.7% · strong
- Index interval
- 10.1–20.1
- Rank posterior
- Not available
- Top 5 / Top 10 probability
- Not available / Not available
- Posterior as of
- Snapshot timestamp unavailable
- Evidence as of
- 13 Sept 2026, 00:00 UTC
- Identity binding
- Exact catalogue product · anthropic/anthropic/claude-3-haiku · aggregates disclosed benchmark configurations · endpoint configuration not claimed
- Index observations used
- 2
- Qualified Index evidence point
- 15.1
Bayesian Evidence Score · measured evidence input only
BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.
- Observed capability
- 15.1
- Evidence status
- confirmed · 77% mass
- General preference
- 15.2
- Coding
- 14.8
- Agents & tools
- —
- Reasoning
- —
- Evidence breadth
- 92%
- Evidence coverage
- 63%
Market and catalogue signals · secondary, never the language Index rank
- Usage (OpenRouter popularity)
- 8.4
- Liquidity (providers × uptime)
- 97.4
- Open reach (HF downloads)
- —
- Surface (context / tools / modalities)
- 63.3
- Freshness
- 3.0
- · OpenRouter popularity #198
- · 1 live provider
- · Catalogue: context 200000, tools, modalities Text/Image
- · First seen 2024-03-13
These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.
BES evidence selected · measured input only
- Arena Overall15.2
- Arena Coding14.8
The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.
Deployment readiness
Metadata index39.0
limited · 56% metadata coverage
- Artifact reproducibility
- 0.0
- Access & legal clarity
- 30.3
- Deployability
- 74.0
- Evaluation provenance
- 53.5
Missing: artifactReproducibility.revision-sha, artifactReproducibility.architecture-and-model-type, artifactReproducibility.parameter-count, artifactReproducibility.serialized-weight-format, artifactReproducibility.base-lineage, accessLegalClarity.declared-license, accessLegalClarity.resolvable-license-terms, evaluationProvenance.evaluation-dataset-revisions.
A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.
Market activity
Venue coverage- Market Gravity
- 36.6
- OpenRouter weekly popularity
- #198 of 322
Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.
Arena preference
- Status
- Ranked
- Source
- LMArena
- Official rank
- #303 of 402
- Rating
- 1261.4
- 95% confidence interval
- 1258–1265
- Votes
- 118K
- Source published at
- 13 Sept 2026, 00:00 UTC
- Category
- overall
- Identity match confidence
- 99%
Market Gravity breakdown
- Usage55%
- 8.4
- Liquidity25%
- 97.4
- Open reach15%
- 50.0
- Freshness5%
- 3.0
OpenRouter weekly popularity across the full model catalogue.
Independent providers versus the model's open or closed cohort, weighted by uptime.
Hugging Face 30-day downloads for open models; neutral for closed models.
Time since first public availability, on a six-month half-life.
Specification
- Context window
- 200K tokens
- Max output
- 4K tokens
- Inputs
- Text, Image
- Outputs
- Text
- Tokenizer
- Claude
- Cached input tokens
- $0.03 / 1M
- First seen on OpenRouter
- 13 Mar 2024
- Knowledge cutoff
- 2023-08-31
Capabilities
- Not supported: Reasoning
- Supported: Tool use
- Not supported: Structured output
- Not supported: Response format
- Supported: Moderated
Benchmark leaderboards featuring Claude 3 Haiku
All leaderboardsClaude 3 Haiku: common questions
How much does Claude 3 Haiku cost per 1M tokens?
Claude 3 Haiku is listed at $0.25 per 1M input tokens and $1.25 per 1M output tokens as of 22 September 2026. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $1.13 total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.
What is the context window of Claude 3 Haiku?
Claude 3 Haiku has a published context window of 200K tokens (200,000) and a maximum output of 4K tokens. Individual providers can serve less than the published maximum; the providers table lists each endpoint's own limit.
Is Claude 3 Haiku open-weight?
API only. No downloadable model weights are linked in the current source data; access is through a hosted API or product.
Which API providers serve Claude 3 Haiku?
One provider serves Claude 3 Haiku through OpenRouter as of 22 September 2026: Amazon Bedrock. Each provider's price, context limit, quantization and measured uptime are in the providers table above.
Which benchmarks has Claude 3 Haiku been evaluated on?
Claude 3 Haiku has published results on 3 public boards tracked by ModelCap as of 22 September 2026: Arena coding 1301 (#296 of 397), AA τ²-Bench Telecom 21.1% (#343 of 436) and AA Humanity's Last Exam 4.1% (#527 of 615). Each board page ranks every tracked model on that benchmark; the ModelCap Index combines them with published uncertainty.
When was Claude 3 Haiku released?
Claude 3 Haiku first appeared in the catalogue on 13 Mar 2024, with a published knowledge cutoff of 2023-08-31. It has since been superseded by Claude Haiku 4.5. Rank and price movements since then are recorded on the site's changes feed.