Context
262K
Max output
256K
Serving providers
5
Cheapest input
$0.20 /1M
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial Analysis and Hugging Face leaderboardsSpeed & latency
Output speed
177 t/s
tokens / second
Time to first token
2.60s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| APEX Agents | 15% | default | — | |||||||||||||||||
| Automation BenchArtificial Analysis did not report this score in its latest run on Oct 14, 2026 | 0.1 | — | — | |||||||||||||||||
| CritPt | 2% | default | — | |||||||||||||||||
| EvalComputeProxyFrozen Aug 18, 2026 | 435.3 | default | — | |||||||||||||||||
| GDPval v2Frozen Aug 18, 2026 | 1017.7 | default | — | |||||||||||||||||
| ||||||||||||||||||||
| GPQA Diamond | 81% | default | — | |||||||||||||||||
| Harvey Lab | 0.7 | default | — | |||||||||||||||||
| Humanity's Last Exam | 21% | default | — | |||||||||||||||||
| IFBench | 67% | default | — | |||||||||||||||||
| IT-Bench SRE | 30% | default | — | |||||||||||||||||
| Long-Context Reasoning | 74% | default | — | |||||||||||||||||
| MMMU-Pro | 75% | default | — | |||||||||||||||||
| Omniscience | -37.3 | default | — | |||||||||||||||||
| Omniscience Accuracy | 0.3 | default | — | |||||||||||||||||
| Omniscience Non Hallucination | 0.2 | default | — | |||||||||||||||||
| SciCode | 44% | default | — | |||||||||||||||||
| SWE-bench Pro | 56% | default | — | |||||||||||||||||
| TerminalBench Hard | 36% | default | — | |||||||||||||||||
| TerminalBench v2.1 | 39% | default | — | |||||||||||||||||
| ||||||||||||||||||||
| τ-bench Banking | 12% | default | — | |||||||||||||||||
| τ²-bench | 99% | default | — | |||||||||||||||||
Artificial Analysis last published scores for this benchmark on Sep 9, 2026. Scores here are frozen at that date, and models released since are not on it.
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.20 to $0.22 per 1M tokens across 5 providers, so the dearest route costs 10% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepInfraCheapest | $0.20 | $1.15 | 1 | |||||||||||||
Standarddeepinfra/stepfun-ai/Step-3.7-Flash
| ||||||||||||||||
| Novita | $0.20 | $1.15 | 1 | |||||||||||||
Standardnovita/stepfun/step-3.7-flash
| ||||||||||||||||
| OpenRouter | $0.20 | $1.15 | 1 | |||||||||||||
Standardstepfun/step-3.7-flash
| ||||||||||||||||
| VercelVercel reseller quote | $0.20 | $1.15 | 1 | |||||||||||||
Standardstepfun/step-3.7-flash
| ||||||||||||||||
| Aihubmix | $0.22 | $1.32 | 1 | |||||||||||||
Standardaihubmix/step-3.7-flash
| ||||||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Cost calculator
estimate your monthly spend on this model$132
estimated / month
Model IDs
copy the exact identifier for your platformstepfun/step-3.7-flashnovita/stepfun/step-3.7-flashdeepinfra/stepfun-ai/Step-3.7-Flashstepfun/step-3.7-flashaihubmix/step-3.7-flash
Frequently asked questions
Step 3.7 Flash pricing, context and availabilityHow much does Step 3.7 Flash cost?
Step 3.7 Flash costs $0.20 per 1M input tokens from DeepInfra, and $1.15 per 1M output tokens from DeepInfra. Across 5 serving providers, input prices range from $0.20 to $0.22 per 1M tokens.
What is the context window of Step 3.7 Flash?
Step 3.7 Flash accepts up to 262K tokens of context and can return up to 256K output tokens.
Which providers serve Step 3.7 Flash?
Step 3.7 Flash is available from 5 serving providers, each with its own pricing and model ID. DeepInfra is currently the cheapest.
What can Step 3.7 Flash do?
Step 3.7 Flash supports image input (vision), tool use, reasoning, prompt caching and structured output.
Other Other models
compare pricing across the Other lineupMuse Glimmer 30B$0.30/1M · 5 providersMuse Spark 1.2$1.25/1M · 4 providersMuse Spark 1.3$1.25/1M · 4 providersMuse Spark 1.2 Contributor$0.10/1M · 3 providersMuse Spark 1.3 Contributor$0.10/1M · 3 providersHy4 preview$0.834/1M · 3 providersMuse Spark 1.1$1.25/1M · 3 providersEmber-1$3.00/1M · 3 providers
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI