Seed OSS 36B Instruct — Benchmarks

Benchmark scores for Seed OSS 36B Instruct aggregated from public leaderboards (3 boards), with how it ranks among open models and how it compares with its nearest neighbours. See hardware requirements for what you need to run it.

llmrun Composite Score →

102.5

90% range 98.1–106.9 · from 3 boards

CodingNot enough data yet
AgentsNot enough data yet
MathNot enough data yet
ScienceNot enough data yet
ReasoningNot enough data yet

Seed OSS 36B Instruct reasoning benchmarks

Multi-step logic and problem-solving — can the model think through a hard problem rather than recall an answer.

GPQA Diamond
GPQA Diamond is a set of extremely hard, graduate-level science questions (physics, chemistry, biology) written by domain experts and filtered so that skilled non-experts with web access still fail. It measures genuine reasoning rather than memorization.
71.5%
71.5%
median 55.7%open leader 93.2%
Overall best: Gemini 4 Argon (high) 96.4%
68th percentile · #33 of 100 open · #179 of 309 all
#33 of 100
#179 of 309
Independent
Epoch AI · Aug 2025

Seed OSS 36B Instruct math benchmarks

Symbolic and quantitative problem-solving, from competition math to multi-step calculation.

AIME 2024/2025
AIME (American Invitational Mathematics Examination) is a prestigious high-school competition of hard, integer-answer problems. It's a widely-cited yardstick for multi-step mathematical reasoning, here on the 2024–2025 papers.
67.5%
67.5%
median 54.4%open leader 99.4%
Overall best: Claude Fable 5 (high) 100.0%
64th percentile · #32 of 88 open · #158 of 292 all
#32 of 88
#158 of 292
Independent
Epoch AI · Aug 2025

Seed OSS 36B Instruct knowledge benchmarks

Breadth and depth of factual knowledge across many subjects, paired with the reasoning to apply it.

MMLU-Pro
MMLU-Pro is a harder, cleaned-up successor to MMLU with ten answer choices and more reasoning-heavy questions across 14 subjects, measuring broad knowledge and reasoning together.
82.7%
82.7%
median 52.0%open leader 88.0%
Overall best: Gemini-3.1-Pro 91.2%
90th percentile · #15 of 146 open · #51 of 257 all
#15 of 146
#51 of 257
Independent
TIGER-Lab

Score vs VRAM in the 24 GB tier

Seed OSS 36B Instruct
Other models
Best Score for its VRAM
x: VRAM (GB) · y: llmrun Score

#8 of 36 that fit 24 GB

Each dot is a model at the best quant that fits a 24 GB GPU. The line steps up to the best Score available at each amount of VRAM; a dot below it is beaten by a model that needs no more memory.

Full 24 GB ranking →

Seed OSS 36B Instruct vs its neighbours

Same boards for the nearest open models per llmrun Score and the best model that fits 24 GB. The highest score in each row is marked.

Board	Seed OSS 36B Instructthis model	Qwen3.8 27Bbest in classQ4_K_M · 18 GB	Gemma 4 26B A4B ITnearest by ScoreQ4_K_M · 17 GB	Qwen3.5 27Bnearest by ScoreQ4_K_M · 18 GB	Qwen3 30B A3B Thinking 2507nearest by ScoreQ4_K_M · 19 GB
llmrun Score	102.5	122.0	105.8	103.8	100.7
GPQA Diamond	71.5%	—	73.2%	—	70.1%
AIME 2024/2025	67.5%	—	82.2%	—	70.3%
MMLU-Pro	82.7%	—	—	86.1%	80.9%
Compare head to head		Compare	Compare	Compare	Compare

Scores aggregated from public benchmark sources (Epoch AI: 2 boards, TIGER-Lab: 1 board) — each linked from the benchmark pages. llmrun does not run these benchmarks. Percentiles and ranks are among open models in the llmrun catalogue; the median and leader marks refer to the same field.