Seed 1.8 Official Specifications, Pricing & Benchmarks
Seed 1.8 (doubao-seed-1-8-251228) by ByteDance 字节豆包, released 2025-12-28, featuring 256,000 context window, 64,000 max output tokens, and Text+Image input.
📋 Seed 1.8 Official API & System Card Specifications
Verified against official ByteDance 字节豆包 developer documentation & pricing
1. Identity & Capabilities
API Spec- API Model ID
- doubao-seed-1-8-251228
- Release Date
- 2025-12-28
- Weight License
- Proprietary (Undisclosed)
- Native Reasoning
- ✅ Supported
- Function Calling & JSON Schema
- ✅ 原生支持
- Batch API 异步批处理
- ✅ Supported (50% off)
2. Context, Output & Modalities
Limits- Max Context Window
- 256,000 tokens (256K)
- Max Output Tokens
- 64,000 tokens
- Input Modalities
- Text + Image
- Output Modalities
- Text
- Direct China Access
- ✅ Direct CN + RMB
- Deployment Mode
- Cloud API only (Proprietary)
3. Official Pricing & Cache
CNY / 1M- Input / 1M Tokens
- ¥ 0.8
- Cached Input / 1M
- ¥ 0.16 -80%
- Output / 1M Tokens
- ¥ 8
🎛️ Official Sampling & Reasoning Parameters
Recommended generation settings & API parameter constraints
📊 Seed 1.8 Full Benchmark Scorecard
56 public benchmarks · ranks among 154 models · synced 2026-11-20
💻 Coding 5 items
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
LiveCodeBench Pro is an advanced evaluation benchmark for large language models for code that uses Elo ratings to rank models based on their performance on coding tasks.
Terminal-Bench 2.0 is an updated benchmark for testing AI agents' tool use ability to operate a computer via terminal.
A multilingual benchmark for issue resolving that evaluates Large Language Models' ability to resolve software issues across diverse programming ecosystems.
AetherCode is a competitive-programming benchmark of olympiad-level algorithmic coding problems.
🤖 Agents & Tools 6 items
A high-difficulty benchmark purpose-built to comprehensively evaluate LLM agents on the Chinese web, consisting of 289 multi-hop questions spanning 11 diverse domains including Film & TV, Technology, Medicine, and History.
AndroidWorld evaluates an agent's ability to operate in real Android GUI environments, completing multi-step tasks by perceiving screen content and executing touch/type actions.
BrowseComp is a benchmark comprising 1,266 questions that challenge AI agents to persistently navigate the internet in search of hard-to-find, entangled information.
WideSearch is an agentic search benchmark that evaluates models' ability to perform broad, parallel search operations across multiple sources. It tests wide-coverage information retrieval and synthesis capabilities.
OSWorld: The first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across Ubuntu, Windows, and macOS with 369 computer tasks involving real web and desktop applications, OS file…
Berkeley Function Calling Leaderboard V4 (BFCL-V4) evaluates LLMs on their ability to accurately call functions and APIs, including simple, multiple, parallel, and nested function calls across diverse programming scenarios.
📐 Math 10 items
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains.
MathVista evaluates mathematical reasoning of foundation models in visual contexts. It consists of 6,141 examples derived from 28 existing multimodal datasets and 3 newly created datasets (IQTest, FunctionQA, and PaperQA), combining challenges from diverse mathematical and…
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks.
MATH-Vision is a dataset designed to measure multimodal mathematical reasoning capabilities.
Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style problem sets.
SuperGPQA is a comprehensive benchmark that evaluates large language models across 285 graduate-level academic disciplines.
A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.
EMMA (Enhanced MultiModal reAsoning) is a benchmark for organic multimodal reasoning across mathematics, physics, chemistry, and coding.
Text-only Humanity's Last Exam variant evaluated with tool use enabled.
👁️ Multimodal 32 items
CountBench evaluates object counting capabilities in visual understanding.
A vision-language benchmark that probes blind spots and brittle reasoning in multimodal models.
AI2D is a dataset of 4,903 illustrative diagrams from grade school natural sciences (such as food webs, human physiology, and life cycles) with over 15,000 multiple choice questions and answers.
The first-ever comprehensive evaluation benchmark of Multi-modal LLMs in Video analysis. Features 900 videos (254 hours) with 2,700 question-answer pairs covering 6 primary visual domains and 30 subfields.
TempCompass is a comprehensive benchmark for evaluating temporal perception capabilities of Video Large Language Models (Video LLMs).
MMMU (Massive Multi-discipline Multimodal Understanding) is a benchmark designed to evaluate multimodal models on college-level subject knowledge and deliberate reasoning.
Video-MMMU evaluates Large Multimodal Models' ability to acquire knowledge from expert-level professional videos across six disciplines through three cognitive stages: perception, comprehension, and adaptation.
MMStar is an elite vision-indispensable multimodal benchmark comprising 1,500 challenge samples meticulously selected by humans to evaluate 6 core capabilities and 18 detailed axes.
A comprehensive benchmark for robust multi-image understanding capabilities of multimodal LLMs. Consists of 12 diverse multi-image tasks involving 10 categories of multi-image relations (e.g., multiview, temporal relations, narrative, complementary).
LiveSports-3K evaluates fine-grained understanding and commentary of live sports video.
LongVideoBench is a question-answering benchmark featuring video-language interleaved inputs up to an hour long.
BLINK: Multimodal Large Language Models Can See but Not Perceive. A benchmark for multimodal language models focusing on core visual perception abilities.
A more robust multi-discipline multimodal understanding benchmark that enhances MMMU through a three-step process: filtering text-only answerable questions, augmenting candidate options, and introducing vision-only input settings.
MMVU (Multimodal Multi-disciplinary Video Understanding) is a benchmark for evaluating multimodal models on video understanding tasks across multiple disciplines, testing comprehension and reasoning capabilities on video content.
LVBench is an extreme long video understanding benchmark designed to evaluate multimodal models on videos up to two hours in duration. It contains 6 major categories and 21 subcategories, with videos averaging five times longer than existing datasets.
OVOBench (Online Video Online Benchmark) evaluates streaming video understanding, testing a model's ability to perceive and respond to video content in real time as it unfolds.
TVBench is a temporal video understanding benchmark evaluating reasoning over actions, events, and temporal dynamics in videos.
CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements.
MotionBench is a benchmark for evaluating multimodal models on motion understanding in videos, testing the ability to comprehend temporal dynamics, movement patterns, and action sequences.
DUDE (Document Understanding Dataset and Evaluation) tests multi-page, multi-domain document understanding and reasoning.
The Abstraction and Reasoning Corpus for Artificial General Intelligence (ARC-AGI) is a benchmark designed to test general intelligence and abstract reasoning capabilities through visual grid-based transformation tasks.
VideoSimpleQA evaluates factual knowledge grounded in video content, measuring how accurately models answer short, fact-seeking questions about videos.
SimpleVQA is a visual question answering benchmark focused on simple queries.
OVBench is an online video understanding benchmark that evaluates a model's ability to perceive, memorize, and reason about real-time video streams as they unfold.
A comprehensive benchmark designed to evaluate image-context reasoning in large visual-language models (LVLMs) by challenging models with 346 images and 1,129 carefully crafted questions to assess language hallucination and visual illusion.
Minerva is a benchmark for complex video reasoning, evaluating models on multi-step reasoning over long and information-dense video content.
VLMsAreBiased evaluates whether vision-language models rely on visual evidence or fall back on language priors when answering.
TOMATO (Temporal Reasoning Multimodal Evaluation) assesses multimodal models on motion and temporal perception in video, testing understanding of actions, motion, and changes over time.
Embodied Reasoning Question Answering benchmark consisting of 400 multiple-choice visual questions across spatial reasoning, trajectory reasoning, action reasoning, state estimation, and multi-view reasoning for evaluating AI capabilities in physical world interactions.
RefSpatialBench evaluates spatial reference understanding and grounding.
ZEROBench is a challenging vision benchmark designed to test models on zero-shot visual understanding tasks.
OmniDocBench 1.5 is a comprehensive benchmark for evaluating multimodal large language models on document understanding tasks, including OCR, document parsing, information extraction, and visual question answering across diverse document types.
🧠 Reasoning & Science 3 items
LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code.
MultiChallenge is a realistic multi-turn conversation evaluation benchmark that challenges frontier LLMs across four key categories: instruction retention (maintaining instructions throughout conversations), inference memory (recalling and connecting details from previous…
PHYBench is a benchmark of real-world physics problems spanning mechanics, electromagnetism, thermodynamics, optics, and modern physics, designed to evaluate physical perception and multi-step quantitative reasoning in large language models.
Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).
Closest Self-Hostable Open-Weight Alternatives to Seed 1.8
Open-weight models with full HuggingFace weights for private GPU / Mac deployment
Frequently Asked Questions
Is Seed 1.8 open-source or proprietary?
Seed 1.8 (doubao-seed-1-8-251228) is a proprietary commercial model by ByteDance 字节豆包. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.
What are the context limit and API pricing of Seed 1.8?
Seed 1.8 supports 256K input context and 64,000 max output tokens, priced at ¥0.8/1M input (¥0.16/1M cached) and ¥8/1M output.