Skip to main content

Step 3.5 Flash

Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token. It is a reasoning model that is incredibly speed efficient even at long contexts.

Input / output price
$0.10 / $0.30 per 1M
Context
262K tokens
Released
Feb 1, 2026
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "stepfun/step-3.5-flash".

Read docs

Providers

3 providers, cheapest first, retired then OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
EmpirioLabsempiriolabs/step-3.5-flash
$0.10/M$0.30/M
StepFunstepfun/step-3.5-flash
$0.10/M$0.30/M
OpenRouterRetiredopenrouter/step-3.5-flash
$0.10/M$0.30/M

Is a provider down? OpenRouter status

Pricing

Averaged across 3 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.10 per 1M
Average output
$0.30 per 1M
Average cache read
$0.02 per 1M

Cost by volume

Total cost from 0 to 1T tokens.

$0.10/MEmpirioLabs, StepFun, and OpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

8.4tokens/sacross providers

Faster than 29% of the other 596 models on the gateway

Response latency

Lower is better

1.4sacross providers

Faster than 53% of the other 758 models on the gateway

Benchmarks

The Respan Index plus 18 published scores from 3 sources, each linked to where it was published.

26.5/ 100

Respan Index#105 of 176

176 ranked models. The shaded band is this model's 95% interval, 16.2 to 48.0. DeepSeek V3.2 and Mistral Medium 3.5 fall inside it, so the gap is within the margin of error.
Step 3.5 Flash by category: Coding 10.6, Agents 22.9, Reasoning 35.8, Knowledge & language 24.1 (estimated), Human preference 44.9
  1. 105Step 3.5 Flash26.5
  2. 104DeepSeek V3.226.6
  3. 106Mistral Medium 3.526.3

Published scores

BenchmarkScore
LiveCodeBench v6static.stepfun.com86.4%
IMOAnswerBenchstatic.stepfun.com85.4%
SWE-bench Verifiedstatic.stepfun.com74.4%
Terminal-Bench 2.0static.stepfun.com51%

Compare Step 3.5 Flash: Step 3.5 Flash vs Step 5 Preview

Apps

Use Step 3.5 Flash in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Step 3.5 Flash FAQ

Short answers, built from the data on this page.

Step 3.5 Flash is a model from StepFun, released on Feb 1, 2026. Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token. It is a reasoning model that is incredibly speed efficient even at long contexts.

On the Respan gateway, Step 3.5 Flash costs $0.10 per 1M input tokens and $0.30 per 1M output tokens on every provider, after gateway discounts.

Step 3.5 Flash is available through 3 providers: EmpirioLabs, StepFun, and OpenRouter.

Step 3.5 Flash accepts up to 262K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "stepfun/step-3.5-flash". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Step 3.5 Flash scores 26.5 out of 100 on the Respan Index v1.1, ranking #105 of 176 models. By category: Coding 10.6, Agents 22.9, Reasoning 35.8, Knowledge & language 24.1 (estimated), and Human preference 44.9. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Step 3.5 Flash has 18 published benchmark scores, including LiveCodeBench v6 86.4%, Tau2-Bench 88.2%, and AIME 2025 98.3%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.