// head_to_head
Claude Haiku 4.5 vs Claude Haiku 5.5
Published LiveBench scores across all seven categories, live list pricing, context windows, and the measured cost of a point of capability — for both models, side by side.
Claude Haiku 5.5 wins outright — it scores higher and costs less.
Claude Haiku 5.5 leads by 28.4 points overall while listing 10× cheaper per blended million tokens. There is no trade-off to reason about here; the only reason to pick the other is a constraint this table does not show, like an existing contract, a region, or a provider you are already on. Claude Haiku 5.5 also leads on measured cost per point of capability, at $0.1429 per point.
anthropic
Claude Haiku 4.5
- Blended / 1M
- $2.00
- Context
- 200K
- Released
- Oct 15, 2025
- Overall score
- 46.0
anthropic
Claude Haiku 5.5
- Blended / 1M
- $0.200
- Context
- 1M
- Released
- Oct 7, 2026
- Overall score
- 74.5
Specs and pricing
| Metric | Claude Haiku 4.5 | Claude Haiku 5.5 |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | 46.0 | 74.5win |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | $0.3893 | $0.1429win |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $2.00 | $0.200win |
| Input price / 1M | $1.00 | $0.100win |
| Output price / 1M | $5.00 | $0.500win |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.100 | $0.010win |
| Context window | 200K | 1Mwin |
| Max output tokens | 64K | 128Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Claude Haiku 4.5 on top, Claude Haiku 5.5 below, both out of 100.
What each one costs to run
Per-token prices are hard to feel. These are monthly list costs for both models across five workload shapes, using each provider's published cached-input rate where there is one.
| Workload | Claude Haiku 4.5 | Claude Haiku 5.5 |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $575.20/mo | $57.52/mo |
| RAG assistant 8K in / 600 out × 100K requests | $740.00/mo | $74.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $696.00/mo | $69.60/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $1,330/mo | $133.00/mo |
| Bulk classification 500 in / 20 out × 5M requests | $2,550/mo | $255.00/mo |
Which should you pick?
You are running this at volume
Claude Haiku 5.5
Lowest measured cost per point of capability at $0.1429 per point — the gap compounds with every request.
Quality matters more than the bill
Claude Haiku 5.5
Highest overall LiveBench score of the two at 74.5.
The workload is coding or agentic work
Claude Haiku 5.5
Leads on coding — 84.5 against 48.1.
You need to fit large documents in one call
Claude Haiku 5.5
Wider context window — 1M against 200K.
Claude Haiku 4.5 vs Claude Haiku 5.5 FAQ
Which is better, Claude Haiku 4.5 or Claude Haiku 5.5?
Claude Haiku 5.5 wins outright — it scores higher and costs less. Claude Haiku 5.5 leads by 28.4 points overall while listing 10× cheaper per blended million tokens. There is no trade-off to reason about here; the only reason to pick the other is a constraint this table does not show, like an existing contract, a region, or a provider you are already on. Claude Haiku 5.5 also leads on measured cost per point of capability, at $0.1429 per point.
Is Claude Haiku 4.5 cheaper than Claude Haiku 5.5?
Claude Haiku 5.5 is cheaper. On a 3:1 input:output blend, Claude Haiku 4.5 lists at $2.00 per million tokens and Claude Haiku 5.5 at $0.200 — Claude Haiku 5.5 is 10× cheaper. Input and output are priced separately — Claude Haiku 4.5 charges $1.00 in and $5.00 out, Claude Haiku 5.5 charges $0.100 and $0.500 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Claude Haiku 4.5 vs Claude Haiku 5.5: which scores higher on benchmarks?
Claude Haiku 4.5 scores 46.0 and Claude Haiku 5.5 scores 74.5 overall on LiveBench, the mean of its seven categories. That is a 28.4-point lead for Claude Haiku 5.5. Category scores differ from the overall figure — a model can lead on reasoning and trail on coding, which the per-category table above breaks out.
Which gives better value for money, Claude Haiku 4.5 or Claude Haiku 5.5?
Claude Haiku 5.5. Cost per point divides the measured dollars LiveBench spent running the benchmark by the score it earned, so it captures something token pricing misses: a reasoning model can emit many times more tokens than its per-token rate suggests. Claude Haiku 4.5 works out at $0.3893 per point and Claude Haiku 5.5 at $0.1429.
Does Claude Haiku 4.5 or Claude Haiku 5.5 have a bigger context window?
Claude Haiku 5.5 has the larger context window: 200K for Claude Haiku 4.5 against 1M for Claude Haiku 5.5. Note that a window you can fill is not a window you should fill — retrieval quality usually degrades well before the limit, and you pay for every token you put in it.
Do Claude Haiku 4.5 and Claude Haiku 5.5 support prompt caching?
Both publish a cached-input rate: $0.100 per million for Claude Haiku 4.5 and $0.010 for Claude Haiku 5.5, against full input rates of $1.00 and $0.100. On a workload with a long stable prefix — a system prompt, a tool schema, a retrieved corpus — that changes the economics more than the headline price does.
Related comparisons
How these numbers are produced
- Price — provider list price from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
- Scores — LiveBench release 2026-09-02, using their own category map. Each model shows its strongest published run. A blank means “not evaluated”, never “bad”.
- Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
- “Win” — awarded only past a threshold: one full point on a benchmark score, 10% on a price, 25% on a context window. Anything tighter reports as a tie, because effort settings alone move a LiveBench score by more than that.
Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.