Step 5 PreviewReady
step-5
Serverless APILLMReasoningVisionPrompt cachingOpen weights
An open-weight challenger with strong benchmark momentum. It ranks #13 of the top 20 frontier models on the Artificial Analysis Intelligence Index with a score of 43.7. Released September 18, 2026.
1M
Context
600B
Parameters
#13
Intelligence rank · 43.7
Open
Weights · StepFun Model License 1.0
View full spec
ReleasedSep 18, 2026
Knowledge cutoff—
Size classlarge
Input modalitiestext, image
Output modalitiestext
Median output speed91 tok/s
Median time to first token2.2s
LicenseStepFun Model License 1.0
Usage guide
1. Choose how you'd like to use Step 5 Preview
☁️ Serverless
Use with pay-per-token pricing. Zero setup, no cold starts, automatic failover between upstream routes.
Usage tiers · Standard • Priority
Try it — sign in
curl https://tokenhub.coality.io/v1/chat/completions \
-H "Authorization: Bearer $TOKEN_HUB_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "step-5",
"messages": [{"role": "user", "content": "Explain vector databases in one paragraph"}]
}'Pricing
| Bucket | Price | Notes |
|---|---|---|
| Input tokens | $1.25 | per 1M tokens |
| Output tokens | $3.38 | per 1M tokens |
| Cache read | $0.1250 | cached input · 90% off |
| Cache write | — | per 1M tokens |
Billed per token with micro-dollar precision · effective Oct 27, 2026 · default key limits 600 req/min · 200K tokens/min
Benchmarks
Humanity's Last Exam
46.5
MMMU-Pro
76.4
SciCode
58.9
Terminal-Bench 4.0
33.3
Long Context Reasoning
88.3
Frontier Code
38.4
Scores from the Artificial Analysis evaluation suite (0–100 scale).
About this model
600B-parameter open-weight model, with a 1M context window, native image input, visible reasoning traces, prompt caching at a 90% discount on cached input, 91 output tokens/s median serving speed and 2.2s to first token.