Anthropic: Claude Haiku 4.5 vs Anthropic: Claude Sonnet 4.5

Side-by-side specs, pricing, and benchmarks. Pick a winner for your team's use case.

Use it in a Space

Spin up a Switchy Space with either model — your whole team @-mentions it with shared context, pooled credits, one memory.

Anthropic: Claude Haiku 4.5

Provider
anthropic
Context
200k
Input $/Mtok
$1.00
Output $/Mtok
$5.00
Max output
64000
Modalities
image, text

Anthropic: Claude Sonnet 4.5

Provider
anthropic
Context
1000k
Input $/Mtok
$3.00
Output $/Mtok
$15.00
Max output
64000
Modalities
text, image, file

Price delta

Anthropic: Claude Haiku 4.5 is $2.00/Mtok cheaper than Anthropic: Claude Sonnet 4.5 on input. Output: Anthropic: Claude Haiku 4.5 is $10.00/Mtok cheaper than Anthropic: Claude Sonnet 4.5.

Which to pick

Pick **Claude Haiku 4.5** for high-frequency, latency-sensitive turns: triage, classification, lightweight transforms, autocomplete-shaped responses. At roughly a quarter of Sonnet's input price ($0.80 vs $3 per Mtok in), it is the right default for any chat where the team is going to send many short turns and the cost compounds. Pick **Claude Sonnet 4.5** when the answer matters more than the response time — drafting longer-form output, reviewing a diff, debating trade-offs in a design conversation. Sonnet's reasoning quality is meaningfully better while still landing well below Opus pricing, which is why most teams keep it as the default Space model and only fall back to Haiku for the cheap-and-fast lanes.
Data last verified 15 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.