Doubao Seed 2.0 Pro
FrontierByteDance•Released on 2026-02-14
ByteDance's flagship foundation model, powering Doubao (China's #1 AI chatbot with 155M weekly users). Achieves frontier-level performance on math (AIME 98.3), coding (Codeforces 3020), and video understanding (VideoMME 89.5). Ranks 6th on LMSYS Text Arena and 3rd on Vision Arena. ~3.7x cheaper than GPT-5.2 on input, ~10x cheaper than Claude Opus 4.5.
85
Overall Score
Core Specs
256K
Context Window
128K
Max Output
✓
Reasoning
✗
Open Source
Multimodal Support
textvisionvideo
Scenario Scores
User Feedback Highlights
Based on community feedback. Hover to see original reviews.
+ Strong agentic capabilities (BrowseComp 77.3)+ ~3.7x cheaper than GPT-5.2, ~10x cheaper than Claude− Limited availability outside China (needs Volcengine account)− Near-zero brand recognition internationally− English optimization secondary to Chinese+ 256K context with multimodal input− Hallucination control weaker than Western models+ Industry-leading video understanding (VideoMME 89.5)− Code generation trails Claude Opus 4.5 (SWE-Bench 76.5 vs 80.9)+ Massive user base and production-hardened+ Frontier math reasoning (AIME 98.3, IMO gold)
Sentiment:👍 72%😐 20%👎 8%
Pros & Cons
Pros
- +Frontier math reasoning (AIME 98.3, IMO gold)
- +Industry-leading video understanding (VideoMME 89.5)
- +~3.7x cheaper than GPT-5.2, ~10x cheaper than Claude
- +Strong agentic capabilities (BrowseComp 77.3)
- +256K context with multimodal input
- +Massive user base and production-hardened
Cons
- −Code generation trails Claude Opus 4.5 (SWE-Bench 76.5 vs 80.9)
- −Hallucination control weaker than Western models
- −English optimization secondary to Chinese
- −Limited availability outside China (needs Volcengine account)
- −Near-zero brand recognition internationally
Pricing
Input (per 1M tokens)$0.47
Output (per 1M tokens)$2.37
Free trial available
Updated on 2026-03-26
Get Started
1Visit the provider's website
2Create an account
3Start using the model
Compare with Others
Benchmarks
userRating8.6%
aime202598.3%
codeforces3020%
gpqaDiamond88.9%
mmluPro87%
sweBenchVerified76.5%
videoMme89.5%
mmmu85.4%