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
Dig into the Haiku 5.5 Benchmark: launch scores, pricing tiers and hard limits, with every figure pinned to its source, config and retrieval date.
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
Reading the Haiku 5.5 Benchmark Without the Hype
Launch-day results, cost tiers and per-effort spend, each one stored alongside its setup details and the date it was pulled.
- What Those Launch Scores Really CaptureElo ratings, partial-credit totals, coding scores and accuracy rates sit on different scales, so averaging them is meaningless.
- Economics of a Small Model at High VolumeDesigned for extraction, labelling and summarisation jobs that run thousands of times a day.
- Two Footnotes That Rewrite the Score TablesThe lowest rates end at 100K tokens, and the 72.4% computer-use figure counts partial credit rather than full passes.
Putting the Haiku 5.5 Benchmark to Work
Three practical steps that convert published Haiku 5.5 Benchmark figures into a real workload decision.
Haiku 5.5 Benchmark — The Findings That Matter
Specifications, cost tiers, coding outcomes, computer-use results and third-party index data for Claude Haiku 5.5, all from published sources.
Context Window and Output Ceiling
Text and images go in, text comes out, and adaptive thinking stays on. A 1M-token context and a 128,000-token output cap make Haiku 4.5's 200,000 and 64,000 look small.
Pricing Splits at the 100K Prompt Mark
Past 100,000 prompt tokens, input, output and caching each move into a tier priced five times higher. OpenAI's base Standard tier runs all the way to 272,000 input tokens.
Terminal-Bench 4.0 Finished Tasks
The 39.2% launch number clears Luna comfortably yet sits below every published Sol row. Each task run allows an eight-hour agent timeout, and any 2.1 or 3.0 figures come from an earlier experiment.
FrontierCode Mergeability and the Cost of Effort
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 stepping from xhigh to max doubles spend for half a point.
OSWorld 2.1: Partial Credit vs. a Strict Pass
The run covers 82 offline tasks with five attempts each, 1080p screenshots and a 500-step cap. When a workflow has to complete, 37.1% under strict grading is the number to trust.
Third-Party Index Scores and Tool Gains
Artificial Analysis lists Haiku across all five effort levels on Intelligence Index v4.3.2, with task costs rounded. HLE shows 45.9% unaided and 57.4% with tools — an uplift that comes from the surrounding system.
Haiku 5.5 Benchmark: Questions Answered
Straight answers on Claude Haiku 5.5 Benchmark results, pricing tiers and hard limits.
Where does Haiku 5.5 stand next to GPT-6 Luna and GPT-6 Sol?
Across knowledge work, screenshot tasks and narrow coding, Haiku and Luna trade wins. Sol and Sol 6.1 charge more and justify it on coding, while Sonnet 5.5 keeps the lead on most rows.
Why doesn't the 72.4% computer-use score count as a success rate?
Require 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. Always add an end-state check.
At what point does Haiku 5.5 move into the higher price tier?
The lowest rates apply only up to 100,000 prompt tokens; beyond that, every category is five times pricier. The 90%, 50% and ~75% savings numbers Anthropic quotes are measured against different baselines.
Could Haiku 5.5 run on local hardware, and how much VRAM would that take?
Published specs give 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 rests on nothing.
Does raising reasoning effort always buy better results?
No. Sol 6.1 actually scores higher at medium than at max, and Haiku's 0.2-point margin over Sonnet at max is far too thin to count as a win. More effort also means more tokens, more latency and more chances to wander off task.
Why does one model show up with different numbers in different tables?
Sonnet shows 70.6% in Anthropic's launch table but 61.8% on the public leaderboard snapshot, where Haiku had no visible row at all. Artificial Analysis removed its 'Default Fallback' label, yet Anthropic's migration guide still states that no server-side fallback exists.
Run the Haiku 5.5 Benchmark Against Your Own Workload
Choose a single workload, fix the effort level, and see whether Haiku 5.5 clears the quality bar you already use.