StepFun launches a $6.99 AI coding plan to rival other open source providers

StepFun launched a new suite of subscription tiers built specifically for high-volume coding and agentic workflows today. The new Step Plans start at an entry-level $6.99 per month and scale up to a $99 flagship tier.

These pricing models directly target independent developers running automated AI agents. Autonomous workflows require massive amounts of continuous API requests. To compete for this audience, rivals like Alibaba and MiniMax also push their own dedicated coding plans.

Still, a closer look at the pricing reveals a massive gap in accessibility.

StepFun: The platform offers four distinct tiers starting at $6.99. The flagship $99 plan grants a rolling limit of 5,000 prompts every five hours.
Alibaba: The company recently discontinued its Lite plan and now forces a flat $50 per month buy-in. This plan imposes a strict hard cap of 90,000 monthly requests.
MiniMax: Tiers range from a $10 entry point up to $150 per month. The company also recently added weekly throttling to restrict continuous operation.

Alibaba’s $50 minimum and strict monthly caps hit developers hard. Autonomous agents burn through multiple requests per single task. That 90,000 limit vanishes incredibly quickly.

Surprisingly enough, the underlying model powering the StepFun plans is flying under the radar. While the community constantly talks about other Chinese models like GLM and Kimi, StepFun’s own Step-3.5-Flash model remains highly underrated.

The OpenRouter surge: Step Flash was recently offered for free on OpenRouter. That exposure quickly made it one of the most popular models on the entire platform.
The performance: The model delivers coding and agentic performance that easily rivals its more famous open-source counterparts.

The Bottom Line: StefFun is still somewhat working from the shadows. The more famous names are grabbing the headlines, but StepFun’s model deliver comparable performance for a better price. The move to pair a highly capable model with a subscription plan that actually tolerates heavy agent workflows, actually positions Step-3.5-Flash as a strong contender among its open-source peers.

Check out more details about Step Plan tiers on Stepfun’s official webiste.

RunPod

If you need on-demand GPUs for training, fine-tuning, inference, or running open-source models, give RunPod a try.

Available hardware: H100, H200, A100, L40S, RTX 4090, RTX 5090, and 30+ more
Cost: significantly cheaper than AWS or GCP, billed per second, no contracts
Setup: spins up in under a minute, 30+ regions worldwide

Try RunPod →

Affiliate disclosure: We may earn a commission if you sign up via our link, at no extra cost to you.