whatmodel.sh / Seed-OSS 36B Instruct
Seed-OSS 36B Instruct
↗ share
+ contribute
Seed-OSS 36B Instruct is an open-weights language model from ByteDance Seed, released August 2025. Its published size is 36.2B parameters and a 524k-token context window. Licence: apache-2.0.
$ how it compares
The LLMs ranked either side of Seed-OSS 36B Instruct, out of 828 with a score.
| # | model | score | where |
|---|---|---|---|
| 195 |  GLM-4.5-Air | 63±1 | api $0.42/M +local |
| 196 |  Qwen3.5 9B | 63±1 | api $0.11/M +local |
| 197 | Seed-OSS 36B Instruct | 63±1 | local |
| 198 |  GPT-4.5 Preview | 63±1 | comparison only |
| 199 |  Trinity-Large-Thinking | 63±2 | api $0.39/M +local |
$ what it's good at
agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #136 of 300 · 15% of its score · 4 tests
Beats DeepSeek V3.2 Exp, Qwen3 235B-A22B Thinking 2507, o4-mini here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Large, GPT-4o (2024-05-13) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWE-bench Verified lab | 56.0% ✎ | ▁▁▁▁▁▁▂▄█▆▁▁ beats 44/131 | 5.0% |
| τ-bench Retail lab | 70.4% ✎ | ▁····▃·▂▇█▃▆ beats 21/28 | 4.2% |
| τ-bench Airline lab | 46.0% ✎ | ▁▁▁·▅▄▁█▇▁▁▇ beats 16/28 | 3.9% |
| Multi-SWE-bench (scaffold per note) lab | 17.0% ✎ | █··█·······█ beats 2/2 | 2.3% |
coding▁▁▁▃▆██▇▆▄▃▁ #122 of 458 · 12% of its score · 3 tests
Beats Qwen3.5 Plus, Gemini 2.5 Pro, Kimi K2 Thinking here, though they rank higher overall. Trails Llama 3.1 Nemotron Ultra 253B v1 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| SWE-bench Verified lab | 56.0% ✎ | ▁▁▁▁▁▁▂▄█▆▁▁ beats 44/131 | 5.0% |
| LiveCodeBench v6 (Feb–May 2025) lab | 67.4% ✎ | ▁▁▃▁█▅▄▇▅▅▃█ beats 18/38 | 4.3% |
| Multi-SWE-bench (scaffold per note) lab | 17.0% ✎ | █··█·······█ beats 2/2 | 2.3% |
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #191 of 393 * · 1.6% of its score · 1 test
Trails LFM2.5 1.2B Instruct, LFM2.5 1.2B Thinking, Gemma 3 12B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| IFEval (aggregation unspecified) lab | 85.8% ✎ | ▁▁▃▁▂▄▄▆█▃▂▁ beats 80/141 | 1.6% |
math▁▂▃▅▆▅▄█▄▁▁▁ #140 of 398 · 20% of its score · 3 tests
Beats Claude Sonnet 4.6, MiMo-V2.5 (310B-A15B), MiniMax-M2.7 here, though they rank higher overall. Trails Olmo 3 32B Think, Olmo 3.1 32B Think, gpt-oss-20b here, though they rank lower overall.
other tests11 not in a task · 58% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| GPQA Diamond +1 | 71.5% ✎ | ▁▁▁▁▂▄▆▄▅▇█▅ beats 108/208 | 26% |
| MMLU lab | 87.4% ✎ | ▁▁▁▁▁▂▃▅█▃▆▃ beats 170/183 | 5.3% |
| Humanity's Last Exam lab | 10.1% ✎ | ▁▁▁▃▆▆▇██▆▁▃ beats 16/104 | 5.2% |
| MMLU-Pro lab | 82.7% ✎ | ▁▁·▁▁▁▃▂▄▅█▁ beats 151/188 | 4.6% |
| MMMLU lab | 78.4% ✎ | ▂▁▁▂▃▂▄▄▅██▃ beats 26/46 | 4.0% |
| SuperGPQA lab | 55.7% ✎ | ▂▁▂▂▅▆▃▇▇▇█▃ beats 27/46 | 3.3% |
| SimpleQA lab | 9.7% ✎ | ▄▄▄▆█▆███·▂█ beats 10/32 | 2.9% |
| KOR-Bench lab | 70.6% ✎ | ▄·······█··▄ beats 2/3 | 2.5% |
3 more tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Chess Puzzles (Epoch) | 13.0% ✎ | █▁▁▁▁▃▁▄▂▂▁▁ beats 64/135 | 2.0% |
| RULER 128k lab | 94.6% ✎ | ▃··▅▄▇▅█▇·▃▃ beats 32/32 | 1.7% |
| BeyondAIME lab | 65.0% ✎ | ▄·▂·▂▄··▄▄▄█ beats 10/15 | 1.0% |
$ where to run it
llama.cpp · recommended
version
- Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.macOS / Linux
curl -LsSf https://llama.app/install.sh | shWindowsirm https://llama.app/install.ps1 | iex - Run Seed-OSS 36B Instruct · needs ~22.22 GB · from huggingface.co/unslothchat in the terminal
llama cli -hf unsloth/Seed-OSS-36B-Instruct-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf unsloth/Seed-OSS-36B-Instruct-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run Seed-OSS 36B Instruct · needs ~22.22 GB · from huggingface.co/unsloth
- Or download it in the terminal
lms get https://huggingface.co/unsloth/Seed-OSS-36B-Instruct-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run Seed-OSS 36B Instruct · needs ~22.22 GB · from huggingface.co/unslothchat in the terminal
ollama run hf.co/unsloth/Seed-OSS-36B-Instruct-GGUF:Q4_K_M
MLX (Mac)
version
- Install MLX (Mac) · Apple silicon Macs only, text only.with uv
uv tool install mlx-lmorpip install mlx-lm - Run Seed-OSS 36B Instruct · needs ~20.3 GB, Mac only · from huggingface.co/lmstudio-communitychat in the terminal
mlx_lm.chat --model lmstudio-community/Seed-OSS-36B-Instruct-MLX-4bit - Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model lmstudio-community/Seed-OSS-36B-Instruct-MLX-4bit
Official weights and documentation: huggingface.co/ByteDance-Seed ↗
$ how the score was worked out · ByteDance-Seed/Seed-OSS-36B-I… 73%, Epoch AI — internal evaluatio… 27%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 ByteDance-Seed/Seed-OSS-36B-Instruct model card · lab-reported | 18 | 73% |
 Epoch AI — internal evaluations | 3 | 27% |
| prior · pull toward the middle | 0% | |
| weighted average | = 62.6 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json