Try MiniMax FastH3 from Fastvideo →
Sign InStart Building

Task History

Click to see output list

No tasks yet

Go to Models
LLM & ChatActive

bytedance / seed-v2-lite

seed-v2-lite

bybytedance

A lightweight ByteDance language model for chat and tool calling, with image understanding and a 256K-token context window.

LLMPartner LLMLLM Tool Call
Model ID
seed-v2-lite
Provider
bytedance
Updated
1788117892
wiro playground—bytedance/seed-v2-lite
Reset to defaults
0 / 20
Maximum 20 image allowed
Delete All

Optional. Multiple image files separated by semicolon for visual understanding.

Required.

Random

Required for chat history. Numeric or string ID.

Random

Required for chat history. Numeric or string ID.

Sample outputs

No samples yet

Run this model to create outputs and build up samples.

Updated 1788117892

Overview

Seed V2 Lite is the lightweight member of ByteDance's Seed 2.0 family. It answers in text, reads images supplied alongside the prompt, and can call tools you define, which suits high-volume work where a larger model would be more than the task needs.

What you can build

  • Chat assistants and support flows that stay responsive under load
  • Tool-using agents that route requests into your own functions
  • Summarisation, classification, and extraction over long documents
  • Visual Q&A on screenshots, charts, and photos

Inputs

  • A written prompt describing the task.
  • Optional images in JPG/JPEG or PNG for visual understanding.
  • Optional system instructions that set role, tone, and boundaries.
  • Optional user and session identifiers, which keep one conversation's history together.
  • Optional sampling controls: temperature, top P, and presence and frequency penalties.
  • Optional reasoning effort, controlling how much the model works through before it answers.

Outputs

A text answer, plus any tool calls the model chooses to make for your client to execute.

Recommended settings

  • Keep temperature near its default for factual work; raise it for drafting and ideation.
  • Raise reasoning effort for multi-step problems, lower it for short and direct answers.
  • Fit the prompt, attached images, and expected answer inside the 256K-token context window.

Limitations

  • This is the lightweight model in the family; hard multi-step reasoning is better served by a larger Seed V2 model.
  • Image understanding depends on input quality. Blurry captures, very small text, and heavy compression hide detail.
  • It returns text only and does not generate images.

Safety & compliance

Follow ByteDance's usage policies for the prompts, images, and generated output you handle, and comply with the laws that apply to your use.

API quick start

Run seed-v2-lite with a single API call.

POST https://api.wiro.ai/v1/Run/bytedance/seed-v2-lite
{
  "prompt": "Tell me about Türkiye",
  "inputImage": "https://your-cdn.com/input.png",
  "user_id": "...",
  "session_id": "..."
}
curl
curl -X POST "https://api.wiro.ai/v1/Run/bytedance/seed-v2-lite" \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_WIRO_API_KEY" \
  --data-binary @- <<'JSON'
{
  "prompt": "Tell me about Türkiye",
  "inputImage": "https://your-cdn.com/input.png",
  "user_id": "...",
  "session_id": "..."
}
JSON

Pricing: Tiered pricing per million tokens. Input: $0.25 (≤128K tokens) / $0.50 (128K–256K tokens). Output: $2.00 (≤128K tokens) / $4.00 (128K–256K tokens).

View full API docs