MODEL PROOF
Step 5 Preview
BETTER VALUE OPTION
A stored alternative has an equal-or-higher measured score and an equal-or-lower price.
Serving precision differs between offers.
MiMo-V2.6-Pro · Δ score 50.3 points · 62% lower measured price · $0.540 / 1M tokens (Price under confirmation since Oct 21, 2026)
GLM 5.3 is a further stored option with these losses: no image, video input
- Independent LMArena score
- 1,440 ± 7.78 · 6,432 votes
- Context window
- 1.05M
- Maximum output
- 64K
- Input modalities
- text, image, video
- Output modalities
- text
- Published input price
- $1.00 / 1M tokens
- Published output price
- $2.70 / 1M tokens
- Pricing kind
- fixed
Inspect complete billing conditions and endpoint terms below.
Price from StepFun: the cheapest offer that is serving, at standard delivery, with at least 90% of the reference context and at the maker’s precision or better, at its standard rate.
Every price dimension · USD per million tokens
| Input | $1.00/M |
|---|---|
| Output | $2.70/M |
| Cached input | $0.100/M |
| Cache write | Unknown/M |
| Cache write 1h | Unknown/M |
| Reasoning | Unknown/M |
Who sells it
The same weights, different shops. Cheapest is not like-for-like when serving precision differs.
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $1.00/M from DeepInfra.
5 provider offers · rates, limits and conditions
Rates are USD per million tokens. Endpoint terms can differ even when the quoted price and precision match. A listed parameter is a provider declaration, not a task-success test.
| Seller | Input | Output | Precision | Uptime (1d) |
|---|---|---|---|---|
| DeepInfradeepinfra/fp8 Endpoint terms
| $1.00 | $2.70 | fp8 | 97.47% |
| StepFunstepfun/fp8 Endpoint terms
| $1.00 | $2.70 | fp8 | 97.62% |
| InferenceNetinference-net/us Endpoint terms
| $1.00 | $2.70 | fp4 | 100.00% |
| InferenceNetinference-net/fp4 Endpoint terms
| $1.00 | $2.70 | fp4 | 100.00% |
| Featherlessfeatherless/fp4 Endpoint terms
| $1.00 | $2.70 | fp4 | 95.24% |
Independent scores
| LMArena Elo | 1440default |
|---|---|
| LMArena row | Step 5 Preview |
LMArena Elo under CC BY 4.0. Absence of another board is not a score of zero.
Capability
| Context window | 1M |
|---|---|
| Max output | 64K |
| Input modes | text, image, video |
| Tool use | yes |
| Reasoning | always on |
| Open weights | Unknown |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-10-21 |
| Quality data | verified |
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional...
Questions this page answers
What does Step 5 Preview cost?
$1.00/M in, $2.70/M out in this catalogue, as of the fetch date on this page. That is the API row, not a subscription.
Does Step 5 Preview have an independent quality score?
Step 5 Preview has an LMArena score in this catalogue.
What beats Step 5 Preview?
MiMo-V2.6-Pro has an equal-or-higher measured score and an equal-or-lower price.
Does this page use Artificial Analysis scores for Step 5 Preview?
No. Artificial Analysis figures are not published here. Quality on this page is LMArena Elo where a score exists; otherwise the row is unrated.
Is the cheapest Step 5 Preview endpoint the same product?
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $1.00/M from DeepInfra.
MODEL MONUMENT