Seed 2.1 Pro
ByteDanceProprietary
विवरण
ByteDance's flagship next-generation agent model built for real-world productivity. A deep-thinking model with strong demand understanding, long-horizon planning, and continuous self-repair, it delivers reliable end-to-end results across complex coding, long-chain agents, and multi-step engineering workflows. Seed 2.1 Pro also advances knowledge, reasoning, and multimodal understanding, with SOTA results across several video understanding benchmarks. Served via Volcano Engine as Doubao-Seed-2.1-pro.
रिलीज़ तिथि
2026-06-24
पैरामीटर
—
संदर्भ लंबाई
—
मोडैलिटीज़
—
क्षमता रडार
80
general
60
coding
70
reasoning
51
scienceअनुमानित
70
agents
70
multimodal
समर्पित विज्ञान बेंचमार्क उपलब्ध न होने पर Science तर्क प्रॉक्सी का उपयोग करके अनुमान लगाता है।
रैंकिंग
| डोमेन | #रैंक | स्कोर | स्रोत |
|---|---|---|---|
| एजेंटिक क्षमता | 38 | 60.0 | LS |
| मल्टीमॉडल रैंकिंग | 70 | 70.0 | LS |
| तर्क | 79 | 56.0 | LS |
बेंचमार्क स्कोर (LLM Stats)
3d
BLINK
81.4%स्वयं
Agents
GDPval
87.9%स्वयं
BrowseComp
86.2%स्वयं
MCP Atlas
83.8%स्वयं
OSWorld
78.8%स्वयं
Web Bench
78.4%स्वयं
MobileWorld
73.1%स्वयं
OfficeQA Pro
72.2%स्वयं
Terminal-Bench 2.1
71.0%स्वयं
CyberGym
70.2%स्वयं
OneMillion Bench
68.8%स्वयं
Agent Startup Bench
68.8%स्वयं
SeedClawBench
66.6%स्वयं
Trae Error Fix
63.3%स्वयं
Trae Code Gen
62.4%स्वयं
WildClawBench
61.7%स्वयं
xDailyBench
61.0%स्वयं
Finance Agent v1.1
60.7%स्वयं
SWE-Bench Pro
57.5%स्वयं
Repo Env
55.0%स्वयं
PresentBench
54.6%स्वयं
Workspace Bench
53.0%स्वयं
Doubao Multi-Turn Bench
52.5%स्वयं
ClawEval-MM
51.0%स्वयं
Toolathlon
50.6%स्वयं
Program Bench
50.3%स्वयं
NL2Repo
47.0%स्वयं
CreativeWork
42.5%स्वयं
Agents' Last Exam
41.4%स्वयं
SWE-Atlas
35.2%स्वयं
APEX-Agents
33.8%स्वयं
DeepSWE
32.7%स्वयं
GameWorld
31.2%स्वयं
PostTrainBench
16.5%स्वयं
Biology
SciCode
59.8%स्वयं
Chemistry
SuperGPQA
70.8%स्वयं
SuperChem
59.8%स्वयं
Code
Artifacts Bench
51.0%स्वयं
FrontierCS
46.3%स्वयं
Coding
AetherCode
65.8%स्वयं
Image2FloorPlan
48.0%स्वयं
Embodied
EmbSpatialBench
0.83 / 100स्वयं
General
MMMU-Pro
82.7%स्वयं
SimpleVQA
0.74 / 100स्वयं
MSQA
50.2%स्वयं
KINA
48.3%स्वयं
Image To Text
OCRBench_V2
63.2%स्वयं
Knowledge
VideoSimpleQA
76.4%स्वयं
WorldBench
67.6%स्वयं
Long Context
DUDE
82.8%स्वयं
LongVideoBench
80.6%स्वयं
MMLongBench-128K
78.3%स्वयं
LVBench
78.0%स्वयं
Math
MathVision
94.5%स्वयं
MathVista
90.7%स्वयं
MathVerse
89.7%स्वयं
Beyond AIME
87.0%स्वयं
EMMA
79.3%स्वयं
FrontierScience Olympiad
75.0%स्वयं
DynaMath
73.1%स्वयं
IMO 2025
0.65 / 42स्वयं
Humanity's Last Exam
55.7%स्वयं
IMOProof-Adv
54.3%स्वयं
MathArena Apex
31.3%स्वयं
LiveMathematicianBench
20.9%स्वयं
HorizonMath
2.0%स्वयं
Multimodal
CharXiv-D
95.5%स्वयं
Video-MME
89.2%स्वयं
CharXiv-R
86.4%स्वयं
VLMsAreBiased
83.6%स्वयं
OVOBench
80.7%स्वयं
TVBench
80.5%स्वयं
TOMATO
79.5%स्वयं
LiveSports-3K
76.8%स्वयं
MotionBench
74.9%स्वयं
BabyVision
73.7%स्वयं
TreeBench
71.1%स्वयं
ChartQAPro
70.9%स्वयं
Minerva
70.7%स्वयं
OVBench
70.0%स्वयं
VideoHolmes
68.2%स्वयं
CrossVid
65.0%स्वयं
ContPhy
63.6%स्वयं
MeasureBench
62.9%स्वयं
ZEROBench
0.56 / 100स्वयं
VisuLogic
0.54 / 100स्वयं
WorldVQA
53.0%स्वयं
VisFactor
51.4%स्वयं
MMSIBench
35.9%स्वयं
Physics
IPhO 2025
79.3%स्वयं
Reasoning
ERQA
72.0%स्वयं
ArcAGI2
62.5%स्वयं
FrontierScience Research
28.3%स्वयं
Spatial Reasoning
RealWorldQA
86.7%स्वयं
AA मूल्यांकन सूचकांक
कोई AA मूल्यांकन डेटा उपलब्ध नहीं
LLM Stats श्रेणी स्कोर
Structured Output100
Search90
Legal80
Long Context80
Spatial Reasoning80
Embodied80
Finance80
General80
3d80
Image To Text70
Math70
Multimodal70
Physics70
Reasoning70
Safety70
Healthcare70
Chemistry70
Economics70
Tool Calling70
Video70
Vision70
Agents60
Biology60
Code60
Frontend Development50
Coding50
Science30
Systems20
मूल्य निर्धारण
कोई मूल्य डेटा उपलब्ध नहीं
गति
कोई गति डेटा उपलब्ध नहीं
प्रदाता मूल्य रैंकिंग
कोई प्रदाता डेटा उपलब्ध नहीं