ByteDance Seed: Seed-2.0-Lite

Public pricingIntelligence 79/100Large memoryVisionDeep thinkingTool use

ByteDance Seed: Seed-2.0-Lite is a multimodal model for agent workflows and tool use. It combines multimodal input handling and reliable tool use and agent behavior with a 262K tokens context window and a low-cost profile. Use it for agent workflows, tool use, and orchestration when latency, cost, and throughput matters.

Input

$0.25/1M

Output

$2.00/1M

Cached

$0.01/1M

Batch

$0.04/1M

Calculate your Seed-2.0-Lite bill.

Set your workload — see cost at your exact volume.

What would Seed-2.0-Lite cost you?

Adjust the workload to see your monthly bill.

1,00010,00050,000250,0001M10M

Technical specifications

Seed-2.0-Lite at a glance.

Memory

262,144

tokens

Max reply

131,072

tokens

Memory tier

Large

an entire book or large codebase

Tokenizer

—

Released

Mar 2026

Training cutoff

Dec 2025

Availability

Public pricing

Status

active

What it can do

Capabilities & limits.

  • Understands images
  • Deep step-by-step thinking
  • Uses tools / calls functions
  • Strict JSON output
  • Streams replies
  • Fine-tunable on your data

When to pick Seed-2.0-Lite

  • Multi-step reasoning, research agents, or hard math.
  • Screenshot analysis, image understanding, or document OCR.
  • Agentic workflows that call tools or APIs.
  • Long documents, full codebases, or extensive chat histories.

When to look elsewhere

  • Very latency-sensitive, real-time apps where every millisecond counts.

Alternatives

Similar models to compare.

FAQ

Seed-2.0-Lite — the questions we see most.

Pricing, capabilities, alternatives — generated from the same data that powers the calculator above.

Get instant answers from our AI agent

At a typical workload of 50,000 conversations a month with 1,500-token prompts and 800-token replies, Seed-2.0-Lite costs roughly $99 per month. Input is $0.25 /1M tokens and output is $2.00 /1M tokens.
Seed-2.0-Lite has a 262,144-token context window (large memory — an entire book or large codebase). That means you can fit about 49,152 words of input and history in a single call.
Beyond text generation, Seed-2.0-Lite supports understanding images, deep step-by-step reasoning, calling functions / tools, strict JSON output. It streams replies by default.
Seed-2.0-Lite was released in March 2026, with training data cut off around December 2025.
Models in a similar class include Seed 1.6, Seed-2.0-Mini, Seed 1.6 Flash. The "Similar models" section below this FAQ links into each.

Still unsure?

Compare Seed-2.0-Lite against 100+ other models.

Open the full wizard — pick a use case, set your usage, and see side-by-side monthly costs in under a minute.

Open the comparison wizard