Seed-OSS-36B-Instruct
ByteDance-Seed/Seed-OSS-36B-Instruct
Official · BF16 · safetensors
Runtime certified: Transformers vLLM
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| BF16 | safetensors | 16 | Official | Runtime certified |
Seed-OSS · General · Seed-OSS 36B
Seed-OSS 36B Instruct: Specs, GPU requirements, hosting & deployment
Seed-OSS 36B Instruct is a current General release in the Seed-OSS family, certified from its immutable official source revision.
Total parameters36.0B
Active parametersDense model
Context window524,288 tokens
Release statusCurrent · Current
ArchitectureSeedOssForCausalLM
ModalitiesText
Licenseapache-2.0
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed
Release date2025-08-20
Architecture, strengths and use cases
Seed-OSS 36B Instruct is an official self-hostable ByteDance Seed release in the Seed-OSS General track. It uses SeedOssForCausalLM, carries 36B, supports a verified context window of 524,288 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
Seed-OSS 36B Instruct uses SeedOssForCausalLM with 36B. The stored architecture profile records 64 layers, hidden width 5120, 80 attention heads and a 524,288 token context ceiling where the official configuration publishes those fields.
Verified SeedOssForCausalLM architecture
524,288 token certified context
Official BF16, Unknown checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Recommended hardware
Use only the listed transformers, vllm runtime configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, runtime overhead, safety headroom, GPU architecture and topology before ranking rentable and reference hardware.
Estimated weight memory69.07 GB
KV cache2 GB
Runtime overhead17.89 GB
Total with safety margin110.31 GB
Recommended best value
2× NVIDIA A100 SXM 80GB
160 GB total VRAM · nvlink
49.69 GB headroom · Available
USD 1,830 / monthLowest-cost viable
5× NVIDIA L4 24GB
120 GB total VRAM · pcie
9.69 GB headroom · Available
USD 1,250 / monthHighest performance
2× NVIDIA A100 SXM 80GB
160 GB total VRAM · nvlink
49.69 GB headroom · Available
USD 1,830 / monthLicense and self-hosting considerations
The full 36B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
Seed-OSS 36B Instruct is published under apache-2.0. The release-scoped official license link and verification date are retained with the catalogue record.
Commercial Use Allowed. The official license permits commercial use; its obligations still apply.
Curated comparisons
Release timeline
bytedance-seed/seed-oss/seed-oss-36b-instruct
Source details
Source revision: c12a95e201ac357725e2665e44ba5e974da39a94fec136d8606138c43fc74699
Frequently asked questions
How much VRAM does Seed-OSS 36B Instruct require?
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible runtime configurations.
Which runtimes support Seed-OSS 36B Instruct?
Certified upstream runtime evidence is available for transformers, vllm.
Can Seed-OSS 36B Instruct be used commercially?
The stored commercial classification is Commercial Use Allowed. Consult the official apache-2.0 terms for the final obligations.
Official references
Last verified: 2026-09-03