Back

ByteDance

Seed-OSS-36B-Instruct

ReasoningOpen weights
Intelligence
12.1
Rank 371 of 744 measured models
Output speed
40.9 t/s
Rank 361 of 403 measured models
Price/1M
$0.30
Rank 206 of 659 measured models
Context window
512K
Rank 343 of 1199 measured models

Seed-OSS-36B-Instruct scores 12.1 on the Artificial Analysis Intelligence Index, ranking 371st of 744 measured models. The closest models on this index are JT-MINI (12.2), Grok 3 (12.1), and Qwen3 235B A22B 2507 Instruct (12.0). At $0.30 per million tokens (3:1 input/output blended price), it is cheaper than Grok 3 ($6) and Qwen3 235B A22B 2507 Instruct ($0.385). With an output speed of 40.9 t/s, it is slower than Grok 3 (100 t/s) and Qwen3 235B A22B 2507 Instruct (43.3 t/s).

Artificial Analysis Indices
Intelligence12.1
Rank 371 of 744 measured models
Math84.7
Rank 48 of 270 measured models
Standard Benchmarks
MMLU Pro81.5%
Rank 86 of 349 measured models
GPQA Diamond72.6%
Rank 298 of 619 measured models
HLE9.9%
Rank 337 of 664 measured models
LiveCodeBench76.5%
Rank 44 of 343 measured models
AIME 202584.7%
Rank 48 of 270 measured models
IFBench41.9%
Rank 254 of 450 measured models
AA-LCR61.3%
Rank 278 of 575 measured models
TerminalBench Hard6.8%
Rank 264 of 432 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy