Loading
90 days of the enterprise, pro and team plans free, no card needed. See the plans
Model comparison
Anthropic's Claude Haiku 5.5 against StepFun's Step 5 Preview: vendor list prices, limits, abilities and the Artificial Analysis indexes side by side - and what OpenSmartRoute does when both are available.
Claude Haiku 5.5 lists at $0.2 per million tokens blended against $1.43 for Step 5 Preview - 7.1x less.
Intelligence index 43 against 44: within a point of each other.
Step 5 Preview reads 1.0M tokens at once, Claude Haiku 5.5 1M.
only Step 5 Preview offers open weights.
Vendor list prices per million tokens and published limits; indexes are Artificial Analysis composites (0-100).
| Specification | Claude Haiku 5.5 | Step 5 Preview |
|---|---|---|
| Input price / 1M tokens | $0.10 | $1.00 |
| Output price / 1M tokens | $0.50 | $2.70 |
| Blended (3 in : 1 out) | $0.20 | $1.43 |
| Cached input / 1M | $0.01 | $0.10 |
| Context window | 1M | 1.0M |
| Max output | 128K | 64K |
| Input | text, image, file | text, image, video |
| Output | text | text |
| Tool calls | ||
| Structured output | ||
| Reasoning | ||
| Open weights | ||
| Intelligence index | 43 | 44 |
| Coding index | - | - |
| Agentic index | - | - |
| Released | Oct 7, 2026 | Oct 8, 2026 |
| Available on this deployment |
List price times tokens for three typical workloads; your mix of input and output decides which model's pricing suits you.
| Workload | Claude Haiku 5.5 | Step 5 Preview | Difference |
|---|---|---|---|
| Support chat20M tokens in, 5M out a month - a help desk answering a few thousand conversations. | $4.50 | $33.50 | $29.00 less on Claude Haiku 5.5 |
| Document summaries100M in, 5M out - long inputs, short answers. | $12.50 | $114 | $101 less on Claude Haiku 5.5 |
| Coding assistant60M in, 20M out - a team's editor assistant, output-heavy. | $16.00 | $114 | $98.00 less on Claude Haiku 5.5 |
The two measure the same on the intelligence index, so the router treats them as peers and lets price, observed latency and your outcomes decide: Claude Haiku 5.5 wins on cost, and a few bad outcomes on either move the next requests to the other.
# targets.yaml - both models as routing targets (metadata.model links each to the catalogue entry)
targets:
- { id: claude-haiku-5.5, kind: llm, metadata: { model: anthropic/claude-haiku-5.5 } }
- { id: step-5-preview, kind: llm, metadata: { model: stepfun/step-5-preview } }
# rules.yaml - the cheaper model by default, the stronger one when the request is hard
rules:
- name: simple-requests-stay-cheap
when: { max_complexity: 0.45 }
prefer: [claude-haiku-5.5]
weight: 0.8
- name: hard-requests-escalate
when: { min_complexity: 0.45 }
prefer: [step-5-preview]
weight: 0.8Every Anthropic and StepFun model, with prices, is on the vendor pages; the rankings show what the router actually picks.