Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Seed 1.8

ByteDance

Strongest in Computer use (#24 of 80, on 1 of its 2 boards), weakest in Agents (#64 of 289). Above par in 2 of 2 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Claude Fable 5.1 and ahead of Gemini 3.7 Flash and EvoCUA.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Agents53.6#64/289
Computer use57.2#24/80

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Seed 1.8, left for the other.

NVIDIA Nemotron 3.5 Lightning 30B A3Bwon 2 · lost 0
Gemma 4 31Bwon 2 · lost 0
NVIDIA Nemotron 3 Ultra 550B A55Bwon 2 · lost 0
Qwen2.5 Vl 32B Instructwon 2 · lost 0
Qwen2.5 Vl 72B Instructwon 2 · lost 0
Kimi Vl A3Bwon 2 · lost 0
Gemini 3.1 Prowon 2 · lost 0
Inklingwon 2 · lost 0
Opencua A3Bwon 2 · lost 0
Mistral Large 4won 2 · lost 0
Opencua Qwen2 7Bwon 2 · lost 0
Claude Haiku 4.5won 2 · lost 0
UI-TARS 72B Dpowon 2 · lost 0
UI-TARS 1.5 7Bwon 2 · lost 0
Opencua 7Bwon 2 · lost 0
Nemotron 3.5 Super VLwon 2 · lost 0
Beam 1.0won 2 · lost 0
TianXi Action 7Bwon 2 · lost 0
Computer Usewon 2 · lost 0
Gui Owl 7Bwon 2 · lost 0
Opencua 32Bwon 2 · lost 0
Claude 3.7 Sonnetwon 2 · lost 0
Doubao 1.5 Thinking Vision Prowon 2 · lost 0
DeepMiner Mano 7Bwon 2 · lost 0
DART GUI 7Bwon 2 · lost 0
Claude Sonnet 4won 2 · lost 0
Qwen3 Vl Flashwon 2 · lost 0
UI-TARSwon 2 · lost 0
MiniMax M3won 2 · lost 0
Opencua 72Bwon 2 · lost 0
Qwen3.8 Flashwon 2 · lost 0
EvoCUA 8Bwon 2 · lost 0
Qwen3.8 27Bwon 2 · lost 0
Muse Spark 1.2won 2 · lost 0
GPT-5.6 Lunawon 2 · lost 0
UI-TARS 2won 2 · lost 0
DeepMiner Mano 72Bwon 2 · lost 0
O3won 2 · lost 0
GPT-5.5won 2 · lost 0
Muse Spark Openwon 2 · lost 0
GUI-Owl-1.5 32Bwon 2 · lost 0
GLM-5.3won 2 · lost 0
Step 5won 2 · lost 0
Grok 4.5won 2 · lost 0
GPT-6 Lunawon 2 · lost 0
EvoCUAwon 2 · lost 0
Gemini 3.7 Flashwon 2 · lost 0
Claude Sonnet 4.6won 1 · lost 1
Kimi K2.5won 1 · lost 1
Kimi K3won 1 · lost 1
Gemini 3.8 Flashwon 1 · lost 1
Qwen3.8 Maxwon 1 · lost 1
DeepSeek V4 Flash Visionwon 1 · lost 1
Claude Sonnet 5won 1 · lost 1
Grok 4.6won 1 · lost 1
DeepSeek V4.1 Flashwon 1 · lost 1
Grok 4.7won 1 · lost 1
Muse Wonderwon 1 · lost 1
Qwen3.7 Pluswon 0 · lost 2
Kimi K2.6won 0 · lost 2
GPT-5.6 Terrawon 0 · lost 2
GLM-5.3 Flashwon 0 · lost 2
GPT-5.6 Solwon 0 · lost 2
Claude Sonnet 4.5won 0 · lost 2
Muse Spark 1.1won 0 · lost 2
Muse Spark 1.3won 0 · lost 2
Holo3 35B A3Bwon 0 · lost 2
Claude Opus 4.8won 0 · lost 2
Mimo V2.6 Prowon 0 · lost 2
GPT-6 Solwon 0 · lost 2
Claude Haiku 5.5won 0 · lost 2
Gemini 4 Argonwon 0 · lost 2
GPT-6.1 Solwon 0 · lost 2
GPT-6 Astrawon 0 · lost 2
Claude Fable 5won 0 · lost 2
Claude Opus 5won 0 · lost 2
Claude Sonnet 5.5won 0 · lost 2
Claude Fable 5.1won 0 · lost 2
Claude Opus 5.5won 0 · lost 2

§ 3 · Sources

Where the numbers come from

1 publication, 1 figures. Every one links to the page it was read from.

OSWorld ↗1 measureread 2026-10-19
OSWorld 61.9%

Badge

PublicAI Index badge for Seed 1.8[![PublicAI Index](https://publicai.io/model-index/badge?model=seed-1-8)](https://publicai.io/model-index/m/seed-1-8)