Available Models
This page contains information on models available in the LLM context.
We provide a collection of open-weight large language models (LLMs) which are self-hosted on our HPC infrastructure, giving you flexibility and choice in your workflows to choose the ones that best fit your needs. User prompts and requests are processed in real time only. The content of these prompts is not saved, logged, or stored at any point. Your data is therefore handled with high protection standards to ensure confidentiality and security.
Model updates
Our list of provied models will be updated in regular intervals, taking several factor into account such as:
- Our available hardware resources for hosting LLMs
- Observed demand for existing and new models
- Features and capabilites of provided and new models
- Usage patterns and utilization
Active models
| Model | Provider | Release Date | max. Content Length | Capabilities | Limitation and Comments |
|---|---|---|---|---|---|
| Mistral-Small-3.2-24B (wird in neuem Tab geöffnet) | Mistral AI | 2025-06-25 | 128K | Compact 24B model optimized for low-latency inference; Good overall performance and quality | |
| Mistral-Small-4-119B-2603 (wird in neuem Tab geöffnet) | Mistral AI | 2026-03-16 | 256K | Mistrals 128 MoE (4 active) model that offers instruct following, reasoning, vision. Designed for general chat assistants, coding, agentic tasks, and reasoning tasks | |
| Apertus-70B (wird in neuem Tab geöffnet) | Swiss-AI | 2025-09-02 | 64K | Strong performance among open models on multilingual / reasoning benchmarks; Medium to good overall quality | Quantized version of Apertus-70B-2509 (wird in neuem Tab geöffnet) with almost the same quality |
| gpt-oss-120B (wird in neuem Tab geöffnet) | OpenAI | 2025-08-06 | 128K | Strong reasoning, coding and benchmark performance; Very good overall performance and quality | |
| Devstral-Small-2-24B-Instruct-2512 (wird in neuem Tab geöffnet) | Mistral AI | 2025-12-09 | 384K | Excellent for coding and agentic workflows | |
| Qwen3-Embedding-8B (wird in neuem Tab geöffnet) | Qwen | 2025-06-05 | 40K | Wide-spread choice for embedding tasks |
Deprecated models
| Model | Provider | Release Date | max. Content Length | Capabilities | Comments |
|---|---|---|---|---|---|
| Mixtral-8x22B (wird in neuem Tab geöffnet) | Mistral AI | 2024-04-17 | 64K | Excels in reasoning, mathematics, coding, multilingual benchmarks |
