Step 3.7 Flash
StepFun · released May 23, 2026 · stepfun-ai/Step-3.7-Flash
- Type
- Open weightsApache License 2.0
- Params
- 201B
- Context
- 262K
active per word not recorded by us · about 197K words of context
Our take
Written Sep 1, 2026Step 3.7 Flash is a 201.4-billion-parameter multimodal model from StepFun with a permissive Apache licence. It accepts text, images and video across a 262,144-token request limit, though no benchmark scores are available to verify its quality.
Pick this for open-weights deployment that needs multimodal inputs with very long context, or for Apache-licensed commercial use where 201-billion-parameter scale is preferred over efficiency. Use it when 85 tokens per second is adequate throughput. Skip it if you need verified quality scores, active-parameter efficiency data, or a lower output rate among permissive alternatives.
The case for it
- 262,144-token request limit among the largest disclosed for this class.
- Apache 2.0 licence with full multimodality: text, image and video in, text out.
- Throughput ranges from 9 to 85 tokens per second across providers at the same rate.
The case against it
- No measured quality scores in our data — chat, reasoning and coding all unverified.
- 201.4 billion total parameters with no active-parameter efficiency disclosed.
- Output rate is steep for an open-weights model with no benchmark to justify the premium over cheaper Apache alternatives.
How good is it?
We hold no score for this model.
So there is no figure here for everyday use, coding, agent work or writing. That is a gap in our data, not a low score.
Can you run it yourself?
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M2 Ultra (76-core GPU) · 192 GB
Borderline fit on an estimated size. It leaves 11.5 GB spare on a size we calculated rather than measured, and a 10% error either way would change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked between 4 hours and 7 days ago — each listing carries its own date.
- per 1M tokens
- $0.20 in / $1.15 out
- Context served
- 262K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $0.20 / $1.15checked 4 hours ago | 262K | not measured | Unknown | Unknown | Unknown |
| DeepInframodeloptDirect | $0.20 / $1.15checked 7 days ago | 262K | not measured | Unknown | Unknown | Unknown |
| Novita AIfp8Direct and through OpenRouter | $0.20 / $1.15checked 4 hours ago | 262K256K max reply through OpenRouter | 17 tok/sthrough OpenRouter | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterConfirmed |
| StepFunfp8Through OpenRouter | $0.20 / $1.15checked 4 hours ago | 256K230K max reply | 103 tok/s | No | Yesunknown period | Unknown |
DeepInfra
Novita AIAcross the 4 listings we hold: 2 say they do not train on prompts (1 of them only through OpenRouter), 0 say they do and 2 do not say. 1 appears in the zero-retention registry we check (only through OpenRouter); the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| DeepInframodeloptDirect | |||
| Novita AIfp8Direct and through OpenRouter | ✓ | ✗ | ✓ |
| StepFunfp8Through OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 3 of 4 listings say yes, 1 publishes no parameter list. JSON output: 2 of 4 listings say yes, 1 says no, 1 publishes no parameter list. Strict schema: 3 of 4 listings say yes, 1 publishes no parameter list.
Models people weigh against Step 3.7 Flash
When we formed this view
Recent changes
What moved
input −20% ($0.20 → $0.16 per 1M tokens), output −20% ($1.15 → $0.92 per 1M tokens), cache read −20% ($0.040 → $0.032 per 1M tokens)What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- No independent board has scored it, so we hold no quality figures at all.
- 1 of 4 listings publishes no parameter list, so what its API accepts is unknown to us.
- We do not hold the active parameter count for it, so how much of it runs on any one token is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 4 listings do not say whether they train on prompts, and 1 answers only through OpenRouter, not for its own listing.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsApache License 2.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- stepfun-ai/Step-3.7-Flash
- Architecture
- Mixture of experts
- Takes in, gives back
- Text, images and video in, text out
- Catalogue slug
- stepfun-step-3-7-flash