ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
ByteDance 字节豆包
Proprietary API ByteDance 字节豆包 · doubao-seed-2-1-pro-260628 · Text+Image+Video

Doubao-Seed-2.1-Pro (豆包旗舰) Official Specifications, Pricing & Benchmarks

ByteDance Seed 2.1 Pro multimodal flagship with 1024K context, 256K output, and low-latency China access.

📋 Doubao-Seed-2.1-Pro (豆包旗舰) Official API & System Card Specifications

Verified against official ByteDance 字节豆包 developer documentation & pricing

API Model ID: doubao-seed-2-1-pro-260628

1. Identity & Capabilities

API Spec
API Model ID
doubao-seed-2-1-pro-260628
Release Date
2026-06-23
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
1,024,000 tokens (1024K)
Max Output Tokens
256,000 tokens
Input Modalities
Text + Image + Video
Output Modalities
Text
Direct China Access
✅ Direct CN + RMB
Deployment Mode
Cloud API only (Proprietary)

3. Official Pricing & Cache

CNY / 1M
Input / 1M Tokens
¥ 6
Cached Input / 1M
¥ 1.2 -80%
Output / 1M Tokens
¥ 30
Tiered Pricing Notes
Starts at ¥0.8/¥2.0 within 32K; tiered up to 1024K context & 256K max output

🏆 Official Benchmark Results

Sourced from official System Card / Model Card & technical reports

Arena Elo ~1492 AA Index 50
📚 Knowledge
MMLU-Pro 87.6 %

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.

🧠 Reasoning
GPQA Diamond 88.5 %

PhD-level Google-proof science reasoning questions

🧠 Reasoning
AIME 2026 93.3 %

All 30 problems from the 2026 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999.

💻 Coding
SWE-bench Verified 76.2 %

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.

👁️ Multimodal
MMMU-Pro 78.8 %

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.

🎛️ Official Sampling & Reasoning Parameters

Recommended generation settings & API parameter constraints

Volcengine Ark Thinking Mode
temperature=1 top_p=0.95 max_tokens=32,768

💡 Supports adaptive thinking toggle (enabled / disabled / auto) in Volcengine Ark API.

📊 Doubao-Seed-2.1-Pro (豆包旗舰) Full Benchmark Scorecard

70 public benchmarks · ranks among 154 models · synced 2026-11-20

⚔️ Compare head-to-head

💻 Coding 13 items

MCP Atlas #6/39 83.8%

MCP Atlas is a benchmark for evaluating AI models on scaled tool use capabilities, measuring how well models can coordinate and utilize multiple tools across complex multi-step tasks.

Terminal-Bench 2.1 #24/37 71%

Terminal-Bench 2.1 is an updated release of the Terminal-Bench benchmark that tests AI agents' ability to operate a computer via the terminal.

CyberGym #12/15 70.2%

CyberGym is a benchmark for evaluating AI agents on cybersecurity tasks, testing their ability to identify vulnerabilities, perform security analysis, and complete security-related challenges in a controlled environment.

AetherCode #1/3 65.8%

AetherCode is a competitive-programming benchmark of olympiad-level algorithmic coding problems.

WildClawBench #2/5 61.7%

WildClawBench is an agentic coding benchmark from InternLM/Claw-Eval that reports overall model performance on real-world tool-using development tasks.

SciCode #2/24 59.8%

SciCode is a research coding benchmark curated by scientists that challenges language models to code solutions for scientific problems.

SWE-Bench Pro #28/52 57.5%

SWE-Bench Pro is an advanced version of SWE-Bench that evaluates language models on complex, real-world software engineering tasks requiring extended reasoning and multi-step problem solving.

Artifacts Bench #2/2 51%

Artifacts Bench evaluates a model's ability to generate visual code artifacts, measuring the quality of generated interactive and visual front-end outputs from natural-language requests.

Program Bench #8/18 50.3%

Program Bench evaluates code-generation agents by asking them to recreate a program's behavior from only a compiled binary and documentation.

NL2Repo #13/23 47%

NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.

SWE-Atlas #1/1 35.2%

SWE-Atlas is a software engineering benchmark focused on debugging, evaluating a model's ability to localize and fix bugs in real-world codebases.

DeepSWE #10/12 32.7%

DeepSWE is a software engineering agent benchmark evaluated with the mini-swe-agent harness, where each task is solved in an isolated container with no internet access. It measures an agent's ability to autonomously resolve real-world coding issues end to end.

PostTrainBench #9/10 16.5%

PostTrainBench evaluates a model's ability to autonomously post-train base models. Given pretrain-only base models, the agent must complete the full pipeline of data synthesis, training, evaluation, and iteration within a time budget, scored across downstream benchmarks such as…

🤖 Agents & Tools 12 items

GDPval #1/3 87.9%

GDPval is an OpenAI benchmark evaluating AI models on economically valuable, real-world knowledge-work tasks spanning many professional occupations and industries.

BrowseComp #12/54 86.2%

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.

OSWorld #1/5 78.8%

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…

MobileWorld #1/1 73.1%

MobileWorld is a benchmark for evaluating multimodal agents on real mobile-device tasks, testing GUI grounding, navigation, and multi-step task completion in mobile environments.

OfficeQA Pro #2/15 72.2%

OfficeQA Pro evaluates AI models on professional knowledge-work questions and tasks drawn from real office workflows, including document analysis, spreadsheet reasoning, and information synthesis across business domains.

OneMillion Bench #1/3 68.8%

OneMillion Bench evaluates AI agents on high-economic-value tasks that require sustained, reliable execution across long-horizon real-world workflows.

Finance Agent v1.1 #1/1 60.7%

Finance Agent v1.1 is an agentic financial-analysis benchmark that evaluates models on real-world finance workflows, including retrieving and reasoning over financial documents and performing multi-step calculations.

Workspace Bench #2/3 53%

Workspace Bench evaluates AI agents on high-economic-value workplace tasks that span multi-step planning, file processing, and tool use across realistic office and productivity workflows.

ClawEval-MM #5/5 51%

ClawEval-MM is the multimodal variant of ClawEval, evaluating agentic problem solving with visual inputs.

Toolathlon #18/32 50.6%

Tool Decathlon is a comprehensive benchmark for evaluating AI agents' ability to use multiple tools across diverse task categories. It measures proficiency in tool selection, sequencing, and execution across ten different tool-use scenarios.

Agents' Last Exam #17/29 41.4%

Agents' Last Exam is a challenging benchmark for AI agents on hard, long-horizon tasks that test sustained reasoning, planning, and tool use, reported with and without tool access.

APEX-Agents #9/15 33.8%

APEX-Agents is a benchmark evaluating AI agents on long horizon professional tasks that require sustained reasoning, planning, and execution across complex multi-step workflows.

📐 Math 10 items

MathVision #6/20 94.5%

MATH-Vision is a dataset designed to measure multimodal mathematical reasoning capabilities.

MathVista #1/7 90.7%

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…

Beyond AIME #1/4 87%

Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style problem sets.

EMMA #1/3 79.3%

EMMA (Enhanced MultiModal reAsoning) is a benchmark for organic multimodal reasoning across mathematics, physics, chemistry, and coding.

FrontierScience Olympiad #1/3 75%

FrontierScience Olympiad is a benchmark of olympiad-level scientific reasoning problems, testing expert understanding and multi-step reasoning across advanced natural-science domains.

DynaMath #6/8 73.1%

A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.

SuperGPQA #4/16 70.8%

SuperGPQA is a comprehensive benchmark that evaluates large language models across 285 graduate-level academic disciplines.

Humanity's Last Exam #11/69 55.7%

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions.

MathArena Apex #6/8 31.3%

MathArena Apex is a challenging math contest benchmark featuring the most difficult mathematical problems designed to test advanced reasoning and problem-solving abilities of AI models. It focuses on olympiad-level mathematics and complex multi-step mathematical reasoning.

HorizonMath #3/4 2%

HorizonMath is an extremely difficult frontier mathematics benchmark designed to test the limits of mathematical reasoning on research-level and competition-beyond problems.

👁️ Multimodal 32 items

CharXiv-D #1/3 95.5%

CharXiv-D is the descriptive questions subset of the CharXiv benchmark, designed to assess multimodal large language models' ability to extract basic information from scientific charts.

Video-MME #1/8 89.2%

Video-MME is the first-ever comprehensive evaluation benchmark of Multi-modal Large Language Models (MLLMs) in video analysis.

RealWorldQA #4/13 86.7%

RealWorldQA is a benchmark designed to evaluate basic real-world spatial understanding capabilities of multimodal models.

CharXiv-R #15/41 86.4%

CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements.

VLMsAreBiased #1/3 83.6%

VLMsAreBiased evaluates whether vision-language models rely on visual evidence or fall back on language priors when answering.

DUDE #1/3 82.8%

DUDE (Document Understanding Dataset and Evaluation) tests multi-page, multi-domain document understanding and reasoning.

MMMU-Pro #6/44 82.7%

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.

BLINK #1/3 81.4%

BLINK: Multimodal Large Language Models Can See but Not Perceive. A benchmark for multimodal language models focusing on core visual perception abilities.

OVOBench #1/3 80.7%

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.

LongVideoBench #1/4 80.6%

LongVideoBench is a question-answering benchmark featuring video-language interleaved inputs up to an hour long.

TVBench #1/4 80.5%

TVBench is a temporal video understanding benchmark evaluating reasoning over actions, events, and temporal dynamics in videos.

TOMATO #1/3 79.5%

TOMATO (Temporal Reasoning Multimodal Evaluation) assesses multimodal models on motion and temporal perception in video, testing understanding of actions, motion, and changes over time.

LVBench #5/15 78%

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.

LiveSports-3K #2/3 76.8%

LiveSports-3K evaluates fine-grained understanding and commentary of live sports video.

VideoSimpleQA #1/3 76.4%

VideoSimpleQA evaluates factual knowledge grounded in video content, measuring how accurately models answer short, fact-seeking questions about videos.

MotionBench #1/4 74.9%

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.

BabyVision #8/15 73.7%

A benchmark for early-stage visual reasoning and perception on child-like vision tasks.

ERQA #4/13 72%

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.

TreeBench #1/2 71.1%

TreeBench evaluates visual grounded reasoning, requiring models to localize and reason about fine-grained visual details.

ChartQAPro #1/3 70.9%

ChartQAPro is a challenging benchmark for question answering over diverse, real-world charts and infographics.

Minerva #1/3 70.7%

Minerva is a benchmark for complex video reasoning, evaluating models on multi-step reasoning over long and information-dense video content.

OVBench #1/3 70%

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.

VideoHolmes #1/3 68.2%

VideoHolmes evaluates video understanding and reasoning capabilities in multimodal models.

CrossVid #1/2 65%

CrossVid evaluates cross-video reasoning, requiring models to integrate information across multiple videos.

ContPhy #1/2 63.6%

ContPhy is a continuum physical-reasoning benchmark evaluating understanding of physical dynamics in video.

OCRBench_V2 #2/5 63.2%

OCRBench v2: Enhanced large-scale bilingual benchmark for evaluating Large Multimodal Models on visual text localization and reasoning with 10,000 human-verified question-answering pairs across 8 core OCR capabilities.

WorldVQA #2/6 53%

WorldVQA is a benchmark designed to evaluate atomic vision-centric world knowledge. It assesses models' ability to understand and reason about visual elements representing real-world knowledge.

VisFactor #1/3 51.4%

VisFactor is a benchmark evaluating fine-grained visual factor perception and reasoning over images.

EmbSpatialBench #5/6 0.8

EmbSpatialBench evaluates embodied spatial understanding and reasoning capabilities.

SimpleVQA #4/13 0.7

SimpleVQA is a visual question answering benchmark focused on simple queries.

VisuLogic #1/2 0.5

VisuLogic evaluates logical reasoning capabilities in visual contexts.

ZEROBench #8/13 0.2

ZEROBench is a challenging vision benchmark designed to test models on zero-shot visual understanding tasks.

Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).

Closest Self-Hostable Open-Weight Alternatives to Doubao-Seed-2.1-Pro (豆包旗舰)

Open-weight models with full HuggingFace weights for private GPU / Mac deployment

Compare with these →

Frequently Asked Questions

Is Doubao-Seed-2.1-Pro (豆包旗舰) open-source or proprietary?

Doubao-Seed-2.1-Pro (豆包旗舰) (doubao-seed-2-1-pro-260628) 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 Doubao-Seed-2.1-Pro (豆包旗舰)?

Doubao-Seed-2.1-Pro (豆包旗舰) supports 1024K input context and 256,000 max output tokens, priced at ¥6/1M input (¥1.2/1M cached) and ¥30/1M output.