whatmodel.sh / Apertus 70B Instruct (2509)
Apertus 70B Instruct (2509)
↗ share
+ contribute
Apertus 70B Instruct (2509) is an open-weights language model from Swiss AI, released September 2025. Its published size is 70.6B parameters and a 66k-token context window. Licence: apache-2.0.
Catalogue identity notes
swiss-ai/Apertus-70B-Instruct-2509 @0f27676 (created 2025-09-01, apache-2.0). Swiss AI Apertus 1.0 70B instruct model, September 2025. Independent rows: EuroEval English (both datasets).
Model information sources
Release date ↗ · Weights licence ↗
HF safetensors metadata 70,599,864,480 parameters.
$ how it compares
The LLMs ranked either side of Apertus 70B Instruct (2509), out of 1132 with a score, then more from Swiss AI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 454 |  Baichuan-M1-14B-Instruct | 54±4 | local | ||
| 455 |  REBEL-Llama-3-Armo-iter_1 | 54±2 | local | ||
| 456 | Apertus 70B Instruct (2509) | 54±6 | local | ||
| 457 |  Haha-7B | 54±6 | local | ||
| 458 |  GPT-4 Turbo Preview (1106) | 54±1 | comparison only | ||
| more from Swiss AI | |||||
| 499 | Apertus 1.5 70B | 53±7 | local | ||
| 650 | Apertus 1.5 8B | 48±7 | local | ||
| 917 | Apertus 8B Instruct (2509) | 38±4 | local | ||
$ what it's good at
other tests2 not in a task · 83% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy | 76.1% ✎ | ▃·▁▁▂▁▂▅█▇▄▁ beats 62/100 | 53% |
| EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy | 88.2% ✎ | ▁▁▁▁▁▁▁▂▄█▄▃ beats 110/146 | 30% |
$ 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 Apertus 70B Instruct (2509) · needs ~43.7 GB · from huggingface.co/unslothchat in the terminal
llama cli -hf unsloth/Apertus-70B-Instruct-2509-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf unsloth/Apertus-70B-Instruct-2509-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run Apertus 70B Instruct (2509) · needs ~43.7 GB · from huggingface.co/unsloth
- Or download it in the terminal
lms get https://huggingface.co/unsloth/Apertus-70B-Instruct-2509-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 Apertus 70B Instruct (2509) · needs ~43.7 GB · from huggingface.co/unslothchat in the terminal
ollama run hf.co/unsloth/Apertus-70B-Instruct-2509-GGUF:Q4_K_M
Other ways to run it
- Run it as published · needs ~141.2 GB, NVIDIA only · from huggingface.co/swiss-ai (official)
vllm serve swiss-ai/Apertus-70B-Instruct-2509
Official weights and documentation: huggingface.co/swiss-ai ↗
$ how the score was worked out · EuroEval leaderboards (result… 83%
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 |
|---|---|---|
 EuroEval leaderboards (results collection), English generative leaderboard | 2 | 83% |
| prior · pull toward the middle | 17% | |
| weighted average | = 54.0 |
$ 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