None
None
Long Story Video Skill
Turn your idea into 10s–10min video—AI writes scripts, prompts, and generates footage automatically.
Ads Video Skill
Generate professional ads and sales videos—AI auto-generates scripts, prompts, and footage.
3D Science Explainer Video Skill
Convert scientific concepts into stunning 3D explain animations
Feedback
freeTrialImage.bannerPity
freeTrialImage.upgradeUnlock
- ✓freeTrialImage.benefitHd
- ✓freeTrialImage.benefitWatermark
- ✓freeTrialImage.benefitUnlimited
Haiku 5.5 Benchmark
Check how the Claude Haiku 5.5 benchmark scores land on coding, agent and knowledge tasks, what each price tier costs and where the hard limits sit.
All Tools
Discover our comprehensive AI-powered animation toolkit
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.
Seedance 2.0
The Future of AI Video Is Here.
Veo3.1
Create Stunning Videos with Veo3.1
FLUX 3 Video Generator
Kling 3.0
Next-Gen AI Video Generator
Grok Video Generator
Create Videos from Text or Images with AI
MiniMax H3 video generator
MiniMax H3 AI Video Generator
Why the Claude Haiku 5.5 Results Deserve a Closer Look
Launch-day scores, price bands and per-effort costs, each stored alongside the configuration and the date it was pulled.
- Which Scale Each Launch Score UsesElo ratings, partial-credit tallies, coding scores and accuracy rates sit on different scales, so averaging them tells you nothing.
- Cost Profile for High-Volume RepetitionDesigned for extraction, labeling and summarization jobs that run thousands of times over.
- Two Footnotes That Rewrite the TablesThe low rate ends at 100K tokens, and the 72.4% computer-use figure counts partial credit rather than full passes.
Using the Haiku 5.5 Benchmark in Three Steps
A three-step method for judging whether published scores match the job you actually run.
Haiku 5.5 Benchmark Highlights: Specs, Pricing and Scores
Specs, price tiers, coding outcomes, computer-use ratings and third-party index results for Claude Haiku 5.5, gathered into one comparison.
Context Window and Output Ceiling
It accepts text plus images and returns text with adaptive thinking enabled. A one-million-token context and a 128,000-token output cap leave Haiku 4.5's 200,000 and 64,000 far behind.
Pricing Split by Prompt Length
Past 100,000 prompt tokens, input, output and caching each move to a tier five times higher. By contrast, OpenAI's base Standard tier stretches to 272,000 input tokens.
Terminal-Bench 4.0: Finished Tasks
Its 39.2% debut clears Luna by a wide margin yet trails the published Sol entries. Runs allow an eight-hour agent timeout, and any 2.1 or 3.0 reading comes from an earlier experiment.
FrontierCode: Mergeability Versus Effort Spend
Scoring weighs correctness, tests, scope, style and repository conventions. At $0.13 per rollout it reaches 46.4% on max effort and 41.6% at medium, yet climbing from xhigh to max doubles cost for half a point.
OSWorld 2.1: Partial Credit Versus Strict Passes
The evaluation covers 82 offline tasks with five attempts each, 1080p screenshots and a 500-step cap. When a workflow has to finish, the 37.1% strict figure is the one to trust.
Third-Party Index Scores and Tool Gains
Artificial Analysis lists it across all five effort settings on Intelligence Index v4.3.2 with rounded task costs. HLE moves from 45.9% to 57.4% once tools are added, an uplift that comes from the surrounding harness.
Haiku 5.5 Benchmark Questions, Answered
Straight answers on Claude Haiku 5.5 scores, token pricing, context limits and local deployment.
How does Haiku 5.5 stack up against GPT-6 Luna and Sol?
Haiku and Luna trade wins across knowledge tasks, screenshot work and focused coding. Sol and Sol 6.1 charge more and justify it on coding, though Sonnet 5.5 still tops most rows.
Why doesn't the 72.4% computer-use figure equal a success rate?
Requiring every checkpoint drops the same run to 37.1%. A model may find and edit the correct file, then save the wrong format or quit before the final step, so always verify the end state.
At what point does Haiku 5.5 move to the pricier tier?
Prompts of 100,000 tokens or fewer keep the lowest rates; beyond that, every category costs five times more. Anthropic's 90%, 50% and ~75% savings numbers rest on different baselines.
Could Haiku 5.5 run on local hardware, and how much VRAM would that take?
The published specs disclose no parameter count, active-parameter figure, layer count or training compute. 'Small' describes a product tier rather than an architecture, so any local GPU estimate lacks a foundation.
Do higher reasoning-effort settings always produce better output?
No. Sol 6.1 scores higher at medium than at max, and Haiku's 0.2-point lead over Sonnet at max is too narrow to count as a win. More effort also means extra tokens, slower responses and more room to drift off task.
Why does one model show different numbers across tables?
Sonnet appears at 70.6% in Anthropic's launch table and 61.8% in the public leaderboard snapshot, where Haiku had no visible row. Artificial Analysis removed its 'Default Fallback' tag, while Anthropic's migration guide still states that no server-side fallback exists.
Measure Haiku 5.5 Against the Work You Actually Run
Choose a single task, dial in the effort level, and see whether it clears the quality bar you set.