Seed OSS 36B Instruct vs GPT OSS 20B

Seed OSS 36B Instruct leads llmrun Score 102.5 to 96.3 (ranges overlap); both fit 24 GB at Q4_K_M; GPT OSS 20B is 6.6× faster on RTX 4090 (~190 vs ~29 tok/s).

Model A · ByteDance-Seed

Seed OSS 36B Instruct

102.5llmrun Score

36.2B · 512K context · Apache 2.0 · Aug 2025

ChatFunctions

Model B · OpenAI

GPT OSS 20B

96.3llmrun Score

20.9B · 4.2B active · 128K context · Apache 2.0 · Aug 2025

ChatFunctions

Which should you choose?

Choose Seed OSS 36B Instruct if you want…

Higher llmrun Score
102.5 vs 96.3, ranges overlap
Ahead on the benchmarks
3 of 3 shared boards
Longer context window
512K vs 128K tokens

Choose GPT OSS 20B if you want…

Fits a 12 GB GPU
IQ4_XS · 11.8 GB; Seed OSS 36B Instruct needs 12.4 GB at its smallest quant
6.6× faster on RTX 4090
~190 vs ~29 tok/s
Lighter on memory
12.9 vs 22.3 GB at Q4_K_M

On your hardware

Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.

MemorySeed OSS 36B InstructGPT OSS 20B

8 GB

RTX 4060 · 8 GB tier

Seed OSS 36B Instruct

Doesn't fit

smallest quant needs 12.4 GB

GPT OSS 20B

IQ2_M · ~103 tok/s

7.6 GB · up to 32Kfits

12 GB

RTX 3060 12GB · 12 GB tier

Seed OSS 36B Instruct

Doesn't fit

smallest quant needs 12.4 GB

GPT OSS 20B

IQ4_XS · ~89 tok/s

11.8 GB · up to 16Kfits

16 GB

RTX 5060 Ti 16GB · 16 GB tier

Seed OSS 36B Instruct

IQ2_M · ~23 tok/s

14.7 GB · up to 16K

GPT OSS 20B

Q4_K_M · ~98 tok/s

13.1 GB · up to 128Kmore headroom

24 GBreference

RTX 4090 · 24 GB tier

Seed OSS 36B Instruct

Q4_K_M · ~29 tok/stight

24.1 GB · up to 8K

GPT OSS 20B

Q4_K_M · ~190 tok/s

13.1 GB · up to 128Kmore headroom

32 GB

RTX 5090 · 32 GB tier

Seed OSS 36B Instruct

Q4_K_M · ~52 tok/s

24.1 GB · up to 64K

GPT OSS 20B

Q4_K_M · ~282 tok/s

13.1 GB · up to 128Kfaster

48 GB

RTX A6000 · 48 GB tier

Seed OSS 36B Instruct

Q4_K_M · ~22 tok/s

24.1 GB · up to 128K

GPT OSS 20B

Q4_K_M · ~154 tok/s

13.1 GB · up to 128Kfaster

Mac 32 GB

Mac Mini M4 · Macs

Seed OSS 36B Instruct

Q4_K_M · ~4 tok/s

24.1 GB · up to 32K

GPT OSS 20B

Q4_K_M · ~28 tok/s

13.1 GB · up to 128Kfaster

Mac 64 GB

Mac Studio M4 Max · Macs

Seed OSS 36B Instruct

Q4_K_M · ~17 tok/s

24.1 GB · up to 256K

GPT OSS 20B

Q4_K_M · ~86 tok/s

13.1 GB · up to 128Kfaster

Mac 128 GB

Mac Studio M4 Max · Macs

Seed OSS 36B Instruct

Q4_K_M · ~17 tok/s

24.1 GB · up to 512K

GPT OSS 20B

Q4_K_M · ~86 tok/s

13.1 GB · up to 128Kfaster

Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).

llmrun Score

with its 90% range · bars share one scale

Seed OSS 36B Instruct GPT OSS 20B

Overall

Seed OSS 36B Instruct

102.598–107

GPT OSS 20B

96.393–99

Coding

Seed OSS 36B Instruct

no data

GPT OSS 20B

100.992–110

Science

Seed OSS 36B Instruct

no data

GPT OSS 20B

92.485–100

Reasoning

Seed OSS 36B Instruct

no data

GPT OSS 20B

91.383–100

Seed OSS 36B Instruct · #46 of 100 open models · #8 of 35 that fit 24 GB

GPT OSS 20B · #63 of 100 open models · #15 of 35 that fit 24 GB

Benchmarks

3 shared boards: Seed OSS 36B Instruct leads 3, GPT OSS 20B leads 0 · Benchmarks imported Oct 2026

BoardSeed OSS 36B InstructGPT OSS 20B

GPQA Diamond

Seed OSS 36B Instruct

71.5%#31/98 open

IndependentEpoch AI · Aug 2025

GPT OSS 20B

60.8%#42/98 open

IndependentEpoch AI · Aug 2025

AIME 2024/2025

Seed OSS 36B Instruct

67.5%#30/86 open

IndependentEpoch AI · Aug 2025

GPT OSS 20B

65.3%#34/86 open

IndependentEpoch AI · Aug 2025

MMLU-Pro

Seed OSS 36B Instruct

82.7%#15/145 open

IndependentTIGER-Lab

GPT OSS 20B

73.6%#33/145 open

IndependentTIGER-Lab

8 more boards (other shared and single-model boards)

Only GPT OSS 20B (8)

LMArena Text1317.4
Terminal-Bench3.4%
SciCode34.4%
WeirdML40.9%
CritPt1.4%
DTBench68.0%
LMCA14.5%
ALE-Bench566.0

VRAM by quantization

Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.

QuantSeed OSS 36B InstructGPT OSS 20B

Q2_K

Seed OSS 36B Instruct

15.9 GB13.6 GB file

32K 20.0 GB · 128K 32.8 GB

GPT OSS 20B

9.3 GB11.5 GB file

32K 9.8 GB · 128K 11.5 GB

Q3_K_S

Seed OSS 36B Instruct

16.4 GB15.9 GB file

32K 20.4 GB · 128K 33.3 GB

GPT OSS 20B

9.5 GB11.5 GB file

32K 10.0 GB · 128K 11.7 GB

Q3_K_M

Seed OSS 36B Instruct

18.2 GB17.6 GB file

32K 22.2 GB · 128K 35.1 GB

GPT OSS 20B

10.6 GB11.5 GB file

32K 11.1 GB · 128K 12.8 GB

Q4_0

Seed OSS 36B Instruct

18.6 GB20.6 GB file

32K 22.7 GB · 128K 35.6 GB

GPT OSS 20B

10.8 GB11.5 GB file

32K 11.3 GB · 128K 13.0 GB

Q4_K_M

Seed OSS 36B Instruct

22.3 GB21.8 GB file

32K 26.3 GB · 128K 39.2 GB

GPT OSS 20B

12.9 GB11.6 GB file

32K 13.4 GB · 128K 15.1 GB

Q5_K_M

Seed OSS 36B Instruct

26.3 GB25.6 GB file

32K 30.4 GB · 128K 43.2 GB

GPT OSS 20B

15.3 GB11.7 GB file

32K 15.8 GB · 128K 17.5 GB

Q6_K

Seed OSS 36B Instruct

30.4 GB29.7 GB file

32K 34.4 GB · 128K 47.3 GB

GPT OSS 20B

17.6 GB12.0 GB file

32K 18.1 GB · 128K 19.8 GB

Q8_0

Seed OSS 36B Instruct

36.7 GB38.4 GB file

32K 40.8 GB · 128K 53.6 GB

GPT OSS 20B

21.3 GB~20.9 GB file

32K 21.8 GB · 128K 23.5 GB

19 more quants

IQ2_XXS

Seed OSS 36B Instruct

10.5 GB10.2 GB file

32K 14.5 GB · 128K 27.4 GB

GPT OSS 20B

6.1 GB~5.8 GB file

32K 6.6 GB · 128K 8.3 GB

IQ2_XS

Seed OSS 36B Instruct

11.4 GB10.9 GB file

32K 15.4 GB · 128K 28.3 GB

GPT OSS 20B

6.7 GB~6.3 GB file

32K 7.2 GB · 128K 8.9 GB

IQ2_S

Seed OSS 36B Instruct

11.9 GB11.6 GB file

32K 15.9 GB · 128K 28.8 GB

GPT OSS 20B

6.9 GB~6.5 GB file

32K 7.4 GB · 128K 9.1 GB

IQ2_M

Seed OSS 36B Instruct

12.8 GB12.7 GB file

32K 16.8 GB · 128K 29.7 GB

GPT OSS 20B

7.4 GB~7.1 GB file

32K 7.9 GB · 128K 9.6 GB

IQ3_XXS

Seed OSS 36B Instruct

14.6 GB14.3 GB file

32K 18.6 GB · 128K 31.5 GB

GPT OSS 20B

8.5 GB11.6 GB file

32K 9.0 GB · 128K 10.7 GB

Q2_K_S

Seed OSS 36B Instruct

15.0 GB12.6 GB file

32K 19.1 GB · 128K 31.9 GB

GPT OSS 20B

8.7 GB12.1 GB file

32K 9.2 GB · 128K 10.9 GB

IQ3_XS

Seed OSS 36B Instruct

15.5 GB15.1 GB file

32K 19.5 GB · 128K 32.4 GB

GPT OSS 20B

9.0 GB11.6 GB file

32K 9.5 GB · 128K 11.2 GB

IQ3_S

Seed OSS 36B Instruct

15.9 GB15.9 GB file

32K 20.0 GB · 128K 32.8 GB

GPT OSS 20B

9.3 GB12.1 GB file

32K 9.8 GB · 128K 11.5 GB

IQ3_M

Seed OSS 36B Instruct

16.8 GB16.5 GB file

32K 20.9 GB · 128K 33.8 GB

GPT OSS 20B

9.8 GB11.6 GB file

32K 10.3 GB · 128K 12.0 GB

Q3_K_L

Seed OSS 36B Instruct

19.1 GB19.1 GB file

32K 23.1 GB · 128K 36.0 GB

GPT OSS 20B

11.1 GB11.5 GB file

32K 11.6 GB · 128K 13.3 GB

IQ4_XS

Seed OSS 36B Instruct

20.0 GB19.5 GB file

32K 24.0 GB · 128K 36.9 GB

GPT OSS 20B

11.6 GB11.6 GB file

32K 12.1 GB · 128K 13.8 GB

IQ4_NL

Seed OSS 36B Instruct

20.9 GB20.6 GB file

32K 24.9 GB · 128K 37.8 GB

GPT OSS 20B

12.1 GB11.6 GB file

32K 12.6 GB · 128K 14.3 GB

Q4_1

Seed OSS 36B Instruct

20.9 GB22.8 GB file

32K 24.9 GB · 128K 37.8 GB

GPT OSS 20B

12.1 GB11.6 GB file

32K 12.6 GB · 128K 14.3 GB

Q4_K_S

Seed OSS 36B Instruct

20.9 GB20.7 GB file

32K 24.9 GB · 128K 37.8 GB

GPT OSS 20B

12.1 GB11.6 GB file

32K 12.6 GB · 128K 14.3 GB

Q4_K_L

Seed OSS 36B Instruct

22.7 GB22.4 GB file

32K 26.7 GB · 128K 39.6 GB

GPT OSS 20B

13.2 GB11.9 GB file

32K 13.7 GB · 128K 15.4 GB

Q5_0

Seed OSS 36B Instruct

23.2 GB25.0 GB file

32K 27.2 GB · 128K 40.1 GB

GPT OSS 20B

not listed

Q5_K_S

Seed OSS 36B Instruct

25.4 GB25.0 GB file

32K 29.5 GB · 128K 42.3 GB

GPT OSS 20B

14.8 GB11.7 GB file

32K 15.3 GB · 128K 17.0 GB

Q5_K_L

Seed OSS 36B Instruct

26.8 GB26.1 GB file

32K 30.8 GB · 128K 43.7 GB

GPT OSS 20B

15.5 GB11.9 GB file

32K 16.0 GB · 128K 17.7 GB

BF16

Seed OSS 36B Instruct

72.9 GB72.3 GB file

32K 76.9 GB · 128K 89.8 GB

GPT OSS 20B

42.2 GB~41.8 GB file

32K 42.7 GB · 128K 44.4 GB

How to run each

Seed OSS 36B Instruct

Ollama · Q4_K_M

$ollama run hf.co/unsloth/Seed-OSS-36B-Instruct-GGUF:Q4_K_M

All ways to run Seed OSS 36B Instruct →

GPT OSS 20B

Ollama · Q4_K_M

$ollama run hf.co/unsloth/gpt-oss-20b-GGUF:Q4_K_M

All ways to run GPT OSS 20B →

Specifications

Specifications of Seed OSS 36B Instruct and GPT OSS 20B
Spec	Seed OSS 36B Instruct	GPT OSS 20B
Publisher	ByteDance-Seed	OpenAI
Parameters	36.2B	20.9B
Active parameters	—	4.2B per token
Architecture	SeedOssForCausalLM	GptOssForCausalLM
Context	512K tokens	128K tokens
Modalities	Chat, Function-calling	Chat, Function-calling
License	Apache 2.0	Apache 2.0
Released	Aug 2025	Aug 2025

Frequently Asked Questions

Which is better, Seed OSS 36B Instruct or GPT OSS 20B?

Seed OSS 36B Instruct leads llmrun Score 102.5 to 96.3 (ranges overlap); both fit 24 GB at Q4_K_M; GPT OSS 20B is 6.6× faster on RTX 4090 (~190 vs ~29 tok/s). Choose Seed OSS 36B Instruct for higher llmrun Score and ahead on the benchmarks; choose GPT OSS 20B for fits a 12 GB GPU and 6.6× faster on RTX 4090.

Which needs less VRAM, Seed OSS 36B Instruct or GPT OSS 20B?

At Q4_K_M, Seed OSS 36B Instruct requires 22.3 GB while GPT OSS 20B needs 12.9 GB, making GPT OSS 20B the easier model to run on your local setup.

Which is faster on RTX 4090?

Estimated decode speed on RTX 4090 is about 29 tok/s for Seed OSS 36B Instruct (Q4_K_M) and 190 tok/s for GPT OSS 20B (Q4_K_M), so GPT OSS 20B is faster. Estimates, ±20%.

Can I run Seed OSS 36B Instruct or GPT OSS 20B on a 16 GB GPU?

Seed OSS 36B Instruct fits at IQ2_M; GPT OSS 20B fits at Q4_K_M.

Which has a longer context window, Seed OSS 36B Instruct or GPT OSS 20B?

Seed OSS 36B Instruct supports 524,288 tokens and GPT OSS 20B supports 131,072 tokens.