ByteDance: Doubao-Seed-2.0-Mini

Lightweight ByteDance Seed 2.0 model for low-latency multimodal reasoning and high-volume tasks

doubao-seed-2.0-mini

AvailableByteDanceChat

Input / Output: $0.10 / $0.40
Context: 229K
Max output: 128K
Modalities: VisionChat
Capabilities: Tool usePrompt CacheReasoning

About Doubao-Seed-2.0-Mini

Doubao-Seed-2.0-Mini is the smallest general model in ByteDance's Seed 2.0 family, optimised for inference throughput and deployment density in high-concurrency scenarios. ByteDance lists efficient batch generation as a focus. It takes text and images, calls tools, and reasons before answering, which suits classification, routing and bulk processing.

Where it works well

Designed by ByteDance for high concurrency, so it is the Seed 2.0 size to use for classification, tagging and routing at volume.
ByteDance reports SWE-Bench Pro 54.4% and Terminal Bench 2.0 60.0% on its Seed2.0 page, notable agentic coding scores for the smallest tier.
Offers tool calling, JSON mode and reasoning, so it can sit inside the same agent code as the larger siblings.
Efficient batch generation is a stated focus, fitting offline jobs such as dataset labelling.

When to choose another model

Complex multi-step reasoning is better served by Seed 2.0 Pro or Lite, which ByteDance positions higher for quality.
Small-model answers on open-ended writing or subtle analysis can be thinner than those of Lite or Pro.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.
02
Send your first request
Copy the example for your language and run it against the endpoint.
Text to textResponses API
POST/v1/responses
API format:
curl https://api.tokenlab.sh/v1/responses \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-xxx" \ -d '{ "model": "doubao-seed-2.0-mini", "input": "Hello!" }'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Input tokens <= 128K · Text to text

per 1M tokens

Official price: Input $0.10 / Output $0.40 / Cache read $0.02 / Cache write $0.10 / Text Input $0.10 / Text Input · Cache read $0.02 / Audio Input $1.50 / Audio Input · Cache read $0.30 / Image Input $0.10 / Image Input · Cache read $0.02 / Text Output $0.40
Verified price: —
Discount: —

Input tokens 128K-256K · Text to text

per 1M tokens

Official price: Input $0.20 / Output $0.80 / Cache read $0.04 / Cache write $0.20 / Text Input $0.20 / Text Input · Cache read $0.04 / Audio Input $3.00 / Audio Input · Cache read $0.60 / Image Input $0.20 / Image Input · Cache read $0.04 / Text Output $0.80
Verified price: —
Discount: —

Prompt cache pricing

Cache read

Official price: $0.02
Verified price: —
Discount: —

Cache write

Official price: $0.10
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Input tokens <= 128K · Text to text	Input $0.10 / Output $0.40 / Cache read $0.02 / Cache write $0.10 / Text Input $0.10 / Text Input · Cache read $0.02 / Audio Input $1.50 / Audio Input · Cache read $0.30 / Image Input $0.10 / Image Input · Cache read $0.02 / Text Output $0.40	—	—
Input tokens 128K-256K · Text to text	Input $0.20 / Output $0.80 / Cache read $0.04 / Cache write $0.20 / Text Input $0.20 / Text Input · Cache read $0.04 / Audio Input $3.00 / Audio Input · Cache read $0.60 / Image Input $0.20 / Image Input · Cache read $0.04 / Text Output $0.80	—	—
Prompt cache pricing
Cache read	$0.02	—	—
Cache write	$0.10	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 17, 2026, 12:52 PM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Doubao-Seed-2.0-Mini in Console with a prompt ready to edit or send.

Help me try doubao-seed-2.0-mini with a short message at /v1/responses. Show the reply, latency, and cost.

Use cases

Best for

Reasoning
Vision

Ticket triage
Label incoming support messages by topic and urgency in a single fast call before a human or larger model sees them.
Bulk content labelling
Run thousands of product images or snippets through the same prompt to produce tags and short descriptions.
Intent routing
Decide in one cheap step which tool, agent or larger model should handle each user request.
Cheap agent sub-steps
Use it for small subtasks inside an agent, such as summarising a tool result, while a larger model plans.

Prompt examples

Classify this message as billing, bug, feature request or other, and give an urgency from 1 to 5. Reply as JSON only.

Write a 12-word alt text for each of these product photos and a comma-separated list of five tags.

Decide which of these three tools the user request needs and output only the tool name.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What exactly is Doubao-Seed-2.0-Mini?

It is the lightweight tier of ByteDance's Seed 2.0 family, built for throughput and dense deployment. It targets high-concurrency work where each call is simple but there are many of them.

What is Mini best used for?

Classification, routing, tagging, extraction and other repeatable tasks run at volume. It also works as a helper inside agents. For difficult reasoning or long autonomous work, step up to Lite or Pro.

Does Doubao-Seed-2.0-Mini accept images?

Yes. It takes text and images, so it can tag photos, read screenshots and caption product shots in bulk. It does not take audio or video.

Does it support reasoning and tool calling?

Yes. A thinking mode and tool calling are both enabled, along with JSON mode and prompt caching, so you can reuse long shared prefixes across many requests.

How does Mini compare with Seed 2.0 Lite?

Mini trades some answer quality for throughput; Lite balances quality and speed and is described as the general production model. Test both on a labelled sample and take Mini if accuracy holds.

How much does Doubao-Seed-2.0-Mini cost?

On TokenLab, Doubao-Seed-2.0-Mini costs Input $0.10 / Output $0.40 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

What are the context window and output limit of Doubao-Seed-2.0-Mini?

Doubao-Seed-2.0-Mini accepts up to 229,376 tokens of context and returns up to 131,072 tokens in one response.

Which endpoint should Doubao-Seed-2.0-Mini use?

Use https://api.tokenlab.sh/v1/responses for Doubao-Seed-2.0-Mini. The request example below shows the matching code shape.

Which operations does Doubao-Seed-2.0-Mini support?

Doubao-Seed-2.0-Mini supports Text to text. Select an operation above to see its endpoint and request example.