RunsOnMy ModelsGPUs & MacsAPI pricesVRAM calculatorNew

Can I run Seed OSS 36B Instruct locally?

Seed OSS 36B Instruct by ByteDance-Seed has 36B parameters. At 4-bit with an 8K context it needs about 23.0 GB of GPU memory. It fits on 36 of 65 GPUs and Macs we track.

23.0 GBQ4_K_M · 8K context
RTX 5090smallest consumer GPU that fits
M4 Pro 48GBsmallest Mac that fits
512Kmax context
GQAattention type

Memory needed by quant and context length

GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from unsloth/Seed-OSS-36B-Instruct-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K) •rough14.4 GB15.4 GB21.4 GB45.4 GB
3-bit (Q3_K_M) •usable18.1 GB19.1 GB25.1 GB49.1 GB
4-bit (Q4_K_M) •good, most popular22.0 GB23.0 GB29.0 GB53.0 GB
5-bit (Q5_K_M) •very good25.5 GB26.5 GB32.5 GB56.5 GB
6-bit (Q6_K) •near lossless29.3 GB30.3 GB36.3 GB60.3 GB
8-bit (Q8_0) •effectively lossless37.5 GB38.5 GB44.5 GB68.5 GB
16-bit (F16/BF16) •original weights69.0 GB70.0 GB76.0 GB100 GB

Which GPUs and Macs run Seed OSS 36B Instruct?

Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).

HardwareMemoryFits?Best quantNeedsMax tok/s (est.)
RTX 3060 12GB12 GBNo———details
RTX 4060 Ti 16GB16 GBNo———details
RTX 309024 GBFitsQ3_K_M19.1 GB53details
RTX 409024 GBFitsQ3_K_M19.1 GB57details
RTX 509032 GBFitsQ5_K_M26.5 GB70details
RTX 5070 Ti16 GBNo———details
RX 7900 XTX24 GBFitsQ3_K_M19.1 GB54details
Mac M4 Pro 48GB48 GBFitsQ5_K_M26.5 GB11details
Mac M4 Max 128GB128 GBFitsF1670.0 GB7.6details
Mac M3 Ultra 512GB512 GBFitsF1670.0 GB11details
DGX Spark128 GBFitsF1670.0 GB3.8details
RTX 407012 GBNo———details
RTX 5060 Ti 16GB16 GBNo———details
RTX 40608 GBNo———details
Arc B58012 GBNo———details
RTX 3060 Ti8 GBNo———details
RTX 30708 GBNo———details
RTX 3080 10GB10 GBNo———details
RTX 3080 12GB12 GBNo———details
RTX 3090 Ti24 GBFitsQ3_K_M19.1 GB57details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBNo———details
RTX 4070 Ti12 GBNo———details
RTX 4070 Ti Super16 GBNo———details
RTX 408016 GBNo———details
RTX 4080 Super16 GBNo———details
RTX 50608 GBNo———details
RTX 507012 GBNo———details
RTX 508016 GBNo———details
2x RTX 30902×24 GBFitsQ8_038.5 GB24details
2x RTX 40902×24 GBFitsQ8_038.5 GB26details
2x RTX 50902×32 GBFitsQ8_038.5 GB47details
RTX A600048 GBFitsQ8_038.5 GB20details
RTX 6000 Ada48 GBFitsQ8_038.5 GB25details
RTX PRO 6000 Blackwell96 GBFitsF1670.0 GB25details
L40S48 GBFitsQ8_038.5 GB22details
A100 40GB40 GBFitsQ6_K30.3 GB52details
A100 80GB80 GBFitsQ8_038.5 GB53details
H100 PCIe80 GBFitsQ8_038.5 GB52details
H100 SXM80 GBFitsQ8_038.5 GB87details
H200141 GBFitsF1670.0 GB66details
RX 76008 GBNo———details
RX 7700 XT12 GBNo———details
RX 7800 XT16 GBNo———details
RX 7900 GRE16 GBNo———details
RX 7900 XT20 GBFitsQ2_K15.4 GB59details
RX 907016 GBNo———details
RX 9070 XT16 GBNo———details
Ryzen AI Max+ 395 128GB128 GBFitsF1670.0 GB3.5details
Arc A770 16GB16 GBNo———details
Mac M1 16GB16 GBNo———details
Mac M2 24GB24 GBTightQ2_K15.4 GB7.4details
Mac M3 24GB24 GBTightQ2_K15.4 GB7.4details
Mac M4 16GB16 GBNo———details
Mac M4 32GB32 GBFitsQ3_K_M19.1 GB6.8details
Mac M1 Pro 32GB32 GBFitsQ3_K_M19.1 GB11details
Mac M3 Pro 36GB36 GBFitsQ3_K_M19.1 GB8.5details
Mac M4 Pro 24GB24 GBTightQ2_K15.4 GB20details
Mac M4 Pro 64GB64 GBFitsQ8_038.5 GB7.1details
Mac M1 Max 64GB64 GBFitsQ8_038.5 GB10details
Mac M2 Max 96GB96 GBFitsQ8_038.5 GB10details
Mac M3 Max 128GB128 GBFitsF1670.0 GB5.5details
Mac M4 Max 64GB64 GBFitsQ8_038.5 GB14details
Mac M2 Ultra 192GB192 GBFitsF1670.0 GB11details
Mac M3 Ultra 256GB256 GBFitsF1670.0 GB11details

Questions

How much VRAM does Seed OSS 36B Instruct need?

About 23.0 GB at 4-bit (Q4_K_M) with an 8K context: 20.3 GB of weights (real GGUF file size), 2.0 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 38.5 GB.

What is the smallest GPU that runs Seed OSS 36B Instruct?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 5090 (32 GB).

Can I run Seed OSS 36B Instruct on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M4 Pro 48GB or larger, using the default macOS GPU memory limit.

Model facts

Hugging Face repoByteDance-Seed/Seed-OSS-36B-Instruct
Parameters36,151,104,512 (36B)
Layers / hidden size64 / 5120
AttentionGQA; 80 heads, 8 KV heads, head dim 128
KV cache per token (FP16)256.0 KiB
Max context524,288 tokens
Licenseapache-2.0
Released on HFAug 20, 2025
Downloads (30d) / likes43,432 / 506

Sources: Hugging Face model API · config.json (ByteDance-Seed/Seed-OSS-36B-Instruct) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.