Seed OSS 36B Instruct
Seed OSS 36B InstructRankingin thinking mode?
Alternatives?
CheaperClaude Haiku 5.5max×3.6-33%$0.21Remove sampling parameters+5
BetterGPT-6 Lunalow+78%-33%$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 27Bxhigh×2.8+219%$1.01Context: 512K → 262K+5
CheaperMiniMax M3.1 Flashthinking×3.5-13%$0.22Complete model response must be appended
CheaperGLM 5.3 FlashmaxOpen×3.5-21%$0.28Memory (4-bit): 22 GB → 193 GB+5Claude Haiku 5.5max×3.6-33%$0.21Remove sampling parameters+5MiniMax M3.1 Flashthinking×3.5-13%$0.22Complete model response must be appendedGLM 5.3 FlashmaxOpen×3.5-21%$0.28Memory (4-bit): 22 GB → 193 GB+5Qwen3.8 FlashthinkingOpen×3.3-23%$0.37Context: 512K → 262K+7GPT-6 Lunamax×3.2-33%$0.068Function calling only with none · Chat Completions+3GPT-6 Lunalow+78%-33%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+74%+50%$0.01
MiMo v2.5thinkingOpen+80%-42%$0.019Memory (4-bit): 22 GB → 186 GBClaude Haiku 5.5low×2.4-33%$0.024Remove sampling parameters+5GPT-5 Minihigh+39%+129%$0.053Context: 512K → 400KQwen3.8 27BxhighOpen×2.8+219%$1.01Context: 512K → 262K+5IK2 Horizon MoVA 36B A4BthinkingOpen×2.1+483%$1 / $4Qwen 3.5 27BthinkingOpen+89%+175%$0.30 / $2.40Context: 512K → 262KAG9v3 39A5BOpen+80%——Context: 512K → 131K+1Qwen3.6 27BthinkingOpen+77%+350%$0.62Context: 512K → 262KDoubao Seed Code+40%——Context: 512K → 256KSeed 2.0 Liten/a—$0.25 / $2Context: 512K → 262KSeed 2.0 Minin/a—$0.10 / $0.40Context: 512K → 262KSeed 2.0 Coden/a—$0.50 / $3Context: 512K → 262KSeed 1.6n/a—$0.25 / $2Context: 512K → 262KSeed 1.6 Flashn/a—$0.075 / $0.30Context: 512K → 262KSeed 2.1 Turbon/a—$0.50 / $2.50Context: 512K → 262KDola Seed 2.1 Turbon/a——Context: not verifiedAll ByteDance Seed models →Run it yourself
estimateBenchmarks
all categoriesFAQ
How much does Seed OSS 36B Instruct cost?›
$0.21 per 1M input tokens and $0.57 per 1M output tokens (cheapest host: SiliconFlow).
Is Seed OSS 36B Instruct free to use?›
There is no free API; the cheapest paid price is $0.21 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Seed OSS 36B Instruct, and what is it best at?›
#193 of 334 in Chat & Knowledge, #142 of 239 in Coding, #109 of 242 in Agents & Tool Use, #144 of 245 in Writing & Language.
Is Seed OSS 36B Instruct better than JT Mini?›
On AA Intelligence Index 4.3.2, Seed OSS 36B Instruct scores 12.1 vs 12.2 — about the same.
How can I use Seed OSS 36B Instruct?›
Through the API of SiliconFlow.
What is the context window of Seed OSS 36B Instruct?›
512K tokens.
When was Seed OSS 36B Instruct released? Is there a newer version?›
Released 2025-08-20. It has been superseded by Doubao Seed Code.
Is Seed OSS 36B Instruct open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Seed OSS 36B Instruct locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Seed OSS 36B Instruct read images?›
No, it accepts text only.
What is Seed OSS 36B Instruct?›
Seed OSS 36B Instruct is an open-weights language model from ByteDance Seed, released Aug 2025.
What are the best alternatives to Seed OSS 36B Instruct?›
Claude Haiku 5.5 (cheaper), GPT-6 Luna (better), Qwen3.8 27B (open weights).