ByteDance: Doubao-Seed-2.1-Lite

Doubao Seed 2.1 Lite supports reasoning across text and images with a large context window. It suits assistants that must connect visual evidence with lengthy instructions, such as reviewing a document alongside screenshots.

Compare models

doubao-seed-2.1-lite

AvailableByteDanceChat

Input / Output: $0.1176 / $0.3971
Context: 1M
Max output: 256K
Modalities: VisionChat
Capabilities: Tool usePrompt CacheReasoning

About Doubao-Seed-2.1-Lite

Doubao-Seed-2.1-Lite is a lighter text-and-image model from ByteDance's Seed 2.1 generation, which ByteDance introduced in June 2026 for coding and agent work. It reasons, calls tools, supports prompt caching, and offers a much longer context than the 2.0 models, which suits assistants that read long documents together with screenshots.

Where it works well

Belongs to the Seed 2.1 generation, which ByteDance built around coding delivery, long agent task chains and stronger multimodal understanding.
Its context window is far larger than the Seed 2.0 models', which helps when a prompt carries whole documents plus screenshots.
Accepts images with text, so it can check a written specification against the screenshots that illustrate it.
Reasoning and tool calling are enabled for assistant and agent use.
Prompt caching is available, which suits repeated long instructions.

When to choose another model

Schema-enforced JSON output is not available, so Seed 2.1 Turbo is the pick when replies must match a schema.
It is the lighter tier of Seed 2.1, so Seed 2.1 Pro suits the hardest coding and agent work better.

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-2.1-lite", "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.

Default spec

per 1M tokens

Official price: Input $0.1176 / Output $0.3971 / Cache read $0.0235
Verified price: —
Discount: —

Prompt cache pricing

Cache read

Official price: $0.0235
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Default spec	Input $0.1176 / Output $0.3971 / Cache read $0.0235	—	—
Prompt cache pricing
Cache read	$0.0235	—	—

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, 12:07 PM UTC

30-day success rate: 99.2%
30-day requests: 100+
Last active: 20 hours ago

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

Open in Console

Open Doubao-Seed-2.1-Lite in Console with a prompt ready to edit or send.

Help me try doubao-seed-2.1-lite with a short message at /v1/responses. Show the reply, latency, and cost.

Use cases

Best for

Vision
Reasoning

Document plus screenshot review
Give it a long specification and the matching UI screenshots and ask where the interface departs from the text.
Long-instruction assistants
Run an assistant whose system prompt carries many policies, relying on caching to reuse that prefix.
Image-aware support bots
Let customers attach screenshots of errors and answer with steps that reference what is visible.
Light agent tasks
Handle tool-using flows that do not need the flagship, such as looking up records and drafting replies.

Prompt examples

The product spec and four screenshots of the build are attached. List every place where the screen disagrees with the spec.

Using the attached policy text, answer the customer's refund question and say which clause you relied on.

The user attached a screenshot of an error dialog. Explain the likely cause and give three steps to try.

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-2.1-Lite?

It is a lower-tier model in ByteDance's Seed 2.1 generation, which ByteDance built around coding delivery, long agent task chains and multimodal understanding. It takes text and images and supports reasoning, tool calling and prompt caching.

Does Seed 2.1 Lite support images?

Yes, it accepts images alongside text and supports reasoning and tool calling. That makes it usable for screenshot review and for documents that contain figures, charts or diagrams.

Does it support JSON schema output?

No. Response-schema output is not supported. Seed 2.1 Turbo and Seed 2.0 Lite both support it, so switch to one of them if a strict schema matters.

How does 2.1 Lite differ from 2.1 Turbo?

Turbo is ByteDance's low-latency production model for scaled agent work and supports schema output. Lite is the lighter tier with a very large context window and no schema enforcement.

Should I choose it over Seed 2.0 Lite?

Choose 2.1 Lite if you need the larger context or want the newer generation. Choose 2.0 Lite if you need JSON schemas or maker-published benchmark results. Compare both on your own prompts.

How much does Doubao-Seed-2.1-Lite cost?

On TokenLab, Doubao-Seed-2.1-Lite costs Input $0.1176 / Output $0.3971 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-2.1-Lite?

Doubao-Seed-2.1-Lite accepts up to 1,048,576 tokens of context and returns up to 262,144 tokens in one response.

Which endpoint should Doubao-Seed-2.1-Lite use?

Use https://api.tokenlab.sh/v1/responses for Doubao-Seed-2.1-Lite. The request example below shows the matching code shape.

Which operations does Doubao-Seed-2.1-Lite support?

Doubao-Seed-2.1-Lite supports Text to text. Select an operation above to see its endpoint and request example.