Monthly cost at list price for five common workloads. Token counts per unit are stated so you can scale them to your own traffic. A workload that does not fit a model's context window is marked rather than priced.
Workload (per month)
Anthropic: Claude Haiku 5.5
Moondream: Moondream 3 Preview
Difference
Customer-support chatbot
50,000 conversations — each a 6-turn conversation (1,500 in / 400 out tokens)
$17.50
$72.50
Anthropic: Claude Haiku 5.5 saves $55.00
RAG search over documents
20,000 questions — each a question answered from ~8 retrieved passages (6,000 in / 500 out tokens)
$17.00
$61.00
Anthropic: Claude Haiku 5.5 saves $44.00
Coding / tool-using agent
2,000 tasks — each a multi-step task with tool calls and file context (60,000 in / 6,000 out tokens)
$18.00
Doesn't fit context
—
Batch summarisation
10,000 documents — each a 6,000-word report summarised to a page (8,000 in / 600 out tokens)
$11.00
$39.00
Anthropic: Claude Haiku 5.5 saves $28.00
Long-document analysis
500 documents — each a 150-page contract or codebase read in one pass (100,000 in / 2,000 out tokens)
$5.50
Doesn't fit context
—
Capabilities side by side
Capability
Anthropic: Claude Haiku 5.5
Moondream: Moondream 3 Preview
What it lets you do
Tool / function calling
✓ Yes
—
call your APIs and run agent loops
Structured output (JSON schema)
✓ Yes
—
return JSON that validates against your schema
Reasoning / thinking mode
✓ Yes
✓ Yes
spend extra tokens thinking before answering hard problems
Image input
✓ Yes
✓ Yes
read screenshots, charts and scanned pages
Audio input
—
—
take speech or audio directly
File / PDF input
✓ Yes
—
accept documents without your own parsing step
Image output
—
—
generate images, not just text
Built-in web search
✓ Yes
—
answer from live web results
Seed (reproducible sampling)
—
—
repeat a generation for tests and evals
Log-probabilities
—
—
read token confidence for classification and scoring
From each model's published parameters and input/output types. “—” means not advertised, not proven absent.
Price history
Anthropic: Claude Haiku 5.5: price unchanged since we started tracking it on 2026-10-08.
Moondream: Moondream 3 Preview: price unchanged since we started tracking it on 2026-10-16.
Which should you choose?
Choose Anthropic: Claude Haiku 5.5 if…
cost matters at scale — it is 76% cheaper on a typical 3:1 input/output mix ($17.50 vs $72.50 a month for 50,000 support chats)
you need to fit long inputs in one request — 1M tokens is roughly 1,500 pages, against 48
you need tool / function calling — to call your APIs and run agent loops; Moondream: Moondream 3 Preview does not advertise it
you need structured output (JSON schema) — to return JSON that validates against your schema; Moondream: Moondream 3 Preview does not advertise it
you need file / PDF input — to accept documents without your own parsing step; Moondream: Moondream 3 Preview does not advertise it
you need built-in web search — to answer from live web results; Moondream: Moondream 3 Preview does not advertise it
you generate long outputs in one go — up to 128K tokens (about 96,000 words) per response
Choose Moondream: Moondream 3 Preview if…
No advantage on the data we track — pick it for provider preference or output quality on your own prompts.
Frequently asked questions
Is Anthropic: Claude Haiku 5.5 better than Moondream: Moondream 3 Preview?
Anthropic: Claude Haiku 5.5 comes out ahead on the data we track: 7 advantages and none for Moondream: Moondream 3 Preview.
Which is cheaper, Anthropic: Claude Haiku 5.5 or Moondream: Moondream 3 Preview?
Anthropic: Claude Haiku 5.5 is cheaper: $0.20 vs $0.85 per 1M tokens at a typical 3:1 input/output mix, 76% less. For 50,000 support conversations a month that is $17.50 against $72.50.
How much do Anthropic: Claude Haiku 5.5 and Moondream: Moondream 3 Preview cost per million tokens?
Anthropic: Claude Haiku 5.5 costs $0.10 per 1M input tokens and $0.50 per 1M output tokens; Moondream: Moondream 3 Preview costs $0.30 input and $2.50 output. Output tokens are billed separately, so long answers shift the gap.
Which has the bigger context window, Anthropic: Claude Haiku 5.5 or Moondream: Moondream 3 Preview?
Anthropic: Claude Haiku 5.5: 1M tokens, about 1,500 pages, against 32K (about 48 pages) for Moondream: Moondream 3 Preview.
Do Anthropic: Claude Haiku 5.5 and Moondream: Moondream 3 Preview support function calling and JSON output?
Tool calling — Anthropic: Claude Haiku 5.5: yes, Moondream: Moondream 3 Preview: not advertised. JSON-schema structured output — Anthropic: Claude Haiku 5.5: yes, Moondream: Moondream 3 Preview: not advertised.
Can Anthropic: Claude Haiku 5.5 or Moondream: Moondream 3 Preview read images?
Yes, both accept image input.
Is Anthropic: Claude Haiku 5.5 or Moondream: Moondream 3 Preview free to use?
No free variant is listed for either; both are pay-per-token.
What is the maximum output length of Anthropic: Claude Haiku 5.5 and Moondream: Moondream 3 Preview?
Anthropic: Claude Haiku 5.5 can write up to 128K tokens (about 96,000 words) per response; Moondream: Moondream 3 Preview up to 4K (about 3,100 words).
When were Anthropic: Claude Haiku 5.5 and Moondream: Moondream 3 Preview released?
Anthropic: Claude Haiku 5.5 was listed on 2026-10-07; Moondream: Moondream 3 Preview on 2026-10-15.
Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...
Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with...