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
See what the Haiku 5.5 Benchmark really shows: launch scores, price tiers and hard limits — all traced to sources, none of it hands-on tested.
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
What the Haiku 5.5 Benchmark Tells You
Every figure traces back to a named evaluation, kept beside its effort setting and the date it was pulled — no blended or averaged scores.
- Scores That Cannot Be MixedElo, partial-credit, coding and accuracy rows sit on different scales, so averaging them yields a meaningless total.
- Cheap Enough for High-Volume JobsDesigned for extraction, tagging and summarising work that repeats thousands of times a day.
- Two Footnotes That Rewrite the TablesDiscount rates end at 100K tokens, and the 72.4% computer-use row is partial credit, not a completed-task rate.
Reading the Haiku 5.5 Benchmark Step by Step
Three moves that turn published Haiku 5.5 Benchmark figures into a real workload decision.
What the Haiku 5.5 Benchmark Data Shows
Specs, both price bands, coding outcomes, computer-use results and third-party index scores for Claude Haiku 5.5, all in one place.
Context Window and Output Ceiling
Text and images in, text out, with adaptive thinking enabled. A 1M-token context and 128K output cap leave Haiku 4.5's 200K and 64K far behind.
Pricing Splits at Two Prompt Lengths
Pass 100,000 prompt tokens and input, output and caching all move to a tier five times dearer. OpenAI's Standard base tier runs to 272,000 input tokens.
Terminal-Bench 4.0 Finished Tasks
At launch Haiku posts 39.2%, clearly ahead of Luna yet below the published Sol rows. Runs allow an eight-hour agent timeout, and any 2.1 or 3.0 figure comes from an earlier experiment.
FrontierCode Mergeability Versus Effort Spent
Scoring weighs correctness, tests, scope, style and repo conventions. Haiku reaches 46.4% at max effort and 41.6% at medium for $0.13 per rollout — yet xhigh to max doubles cost for half a point.
OSWorld 2.1: Partial Credit or Strict Pass
The run covers 82 offline tasks, five attempts each, 1080p screenshots and a 500-step cap. If a workflow must finish, the 37.1% strict rate is the number to trust.
Third-Party Index Scores and Tool Gains
Artificial Analysis lists Haiku at all five efforts on Intelligence Index v4.3.2 with rounded task costs. HLE shows 45.9% unaided and 57.4% with tools — a lift from the harness, not the model alone.
Haiku 5.5 Benchmark: Questions Answered
Straight answers on Claude Haiku 5.5 Benchmark scores, price tiers and published limits.
Where does Haiku 5.5 sit next to GPT-6 Luna and GPT-6 Sol?
Haiku trades blows with Luna on knowledge work, screenshot tasks and narrow coding. Sol and Sol 6.1 charge more and justify it on coding, while Sonnet 5.5 still tops most rows.
Why can't the 72.4% computer-use figure be called a success rate?
Demand every checkpoint and the same run drops to 37.1%. An agent may find and edit the right file, then save it in the wrong format or quit before the final action. Add an end-state check.
At what point does Haiku 5.5 move to the pricier band?
Prompts up to 100,000 tokens are the only ones billed at the lowest rates; beyond that line every category costs five times more. Anthropic's 90%, 50% and ~75% savings figures each use a different baseline.
Could Haiku 5.5 run locally, and what GPU memory would that take?
The published specs give no parameter count, active-parameter figure, layer count or training compute. 'Small' describes a product tier rather than an architecture, so local GPU estimates have nothing to stand on.
Does extra reasoning effort always pay off?
Not always. Sol 6.1 scores better at medium than at max, and Haiku's 0.2-point edge over Sonnet at max is too small to call a win. More effort means more tokens, more latency and more chances to drift off task.
Why do the same models post different numbers in different tables?
Sonnet reads 70.6% in Anthropic's launch table but 61.8% on the public leaderboard snapshot, where Haiku had no visible row. Artificial Analysis retired its 'Default Fallback' label, yet Anthropic's migration guide still says no server-side fallback exists.
Put the Haiku 5.5 Benchmark to Work on Your Own Tasks
Choose a single task, dial in the effort level, and see whether Haiku 5.5 clears the quality bar you already use.