ByteDance Seed 2.0 Mini
The smallest, cheapest Seed 2.0 model for fast everyday tasks, with tool calling and image input.
Seed 2.0 Mini is a cloud text model from ByteDance. It is multimodal: alongside a text prompt it accepts images, then replies with generated text. Its context window handles up to 262K input tokens and up to 131K output tokens, and it supports adjustable reasoning effort for harder problems. Generation can be tuned with system instructions, temperature, and top-p sampling. It runs through OpenRouter and ByteDance using your own API key, from $0.10 per million input tokens.
- Released
- 5 months ago (Apr 2026)
- Pricing
- $0.10 / 1M in · $0.40 / 1M out
- Inputs
- Text, Images
- Context window
- 262K in · 131K out
- Reasoning
- Adjustable effort
- Controls
- System prompt, Temperature, Top-p
Examples
Generated with Seed 2.0 Mini via OpenRouter. The same three prompts run against every text model in the catalog, shown verbatim, so the only thing that changes between two models’ answers is the model.
Explain what an API is to someone who has never written code. Do not use a restaurant or waiter analogy. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 50 words.
An API is a set of clear rules that lets different apps or programs share information and work together smoothly, so one tool can send your data to another without extra hassle.
A two-person startup is deciding whether to build their own user authentication or use a hosted provider. Give them a clear recommendation, then the single strongest argument against your own recommendation. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 150 words.
My recommendation is to use a hosted user authentication provider like Auth0 or Firebase Auth. For a two-person startup, building in-house would drain critical engineering time that should go to product features driving user growth, not security infrastructure. Hosted solutions set up fast, handle compliance and maintenance automatically, freeing your small team to focus on what makes your product unique. The single strongest argument against this is that hosted providers create a risky third-party dependency: a major outage would cut off all user logins, alienate customers, and harm your brand’s reputation.
Write a briefing for a non-technical executive explaining why their company's spending on AI is hard to predict, and what they can do about it. Cover at least three distinct causes. Plain prose only — no headings, bullet points, or markdown formatting. Write exactly 205 words.
As you’ve seen, our company’s AI spending is hard to pin down because of three distinct factors that don’t align with traditional budgeting for tools or services. First, AI models and workflows evolve so quickly that we often pivot to a better-performing open-source model or fine-tune existing ones to fit our unique business needs, which can add unplanned costs for data processing or specialized technical expertise we didn’t budget for at project kickoff. Second, most AI vendors charge usage-based fees tied to calls, data storage, or compute power, so when our teams test new features or scale projects to meet growing demand, those costs can jump unexpectedly—even if we thought we fully planned for all related expenses. Third, integrating AI with our legacy systems often requires extra support from third-party consultants or stretched internal IT teams that we didn’t account for in initial budget estimates. To address this, we can adopt rolling monthly AI cost tracking, set aside a small dedicated contingency fund for unplanned experimentation, and negotiate fixed-price pilot agreements with vendors for new tools to limit future surprises. This approach keeps our spending aligned with strong clear project outcomes while allowing us to leverage AI’s key competitive benefits without overextending our overall budget.
ByteDance
ByteDance's AI labs ship the Seedream image models and the Seedance video models, among other Doubao-branded research.
www.bytedance.com ↗