Leaderboard / OpenGPT-X (Teuken)
Teuken 7B Instruct v0.6
OpenGPT-X (Teuken) · Teuken · released 2025-07-28 · data checked 2026-11-03
Teuken 7B Instruct v0.6 is a 7.5B-parameter dense model from OpenGPT-X (Teuken) (Europe), released 2025-07-28 under the CC-BY-NC-4.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 5.3 GB of memory at 4-bit with an 8K-token context.
Benchmarks
Each bar shows this model's score. The tick marks the best small model on that benchmark. Sources: 1 reported by OpenGPT-X (Teuken). How we label sources.
Earlier version: Teuken 7B Instruct R v0.4 Earlier version
Published for the previous release, not this one. Shown for context only: they don't count toward the index or appear in tables. Measured by another lab. Source: huggingface.co, arxiv.org.
Specifications
- Lab
- OpenGPT-X (Teuken) (DE · Europe)
- Architecture
- Dense transformer
- Total parameters
- 7.5B
- Active parameters
- All (dense)
- Size class
- Small
- Context window
- 4K tokens
- Input modalities
- text
- Reasoning
- Instruct (no thinking)
- License
- CC-BY-NC-4.0
- Openness
- Open weights
- Memory to run
- ~5.3 GB at 4-bit, 8K context (more sizes)
- Hugging Face
- 1.1K downloads (30 days) · 18 likes
Compare with
- Teuken 7B Instruct v0.6 vs Hunyuan 7B Instruct (7.5B, index 52.5)
- Teuken 7B Instruct v0.6 vs Falcon H1R 7B (7.6B, index 57.5)
- Teuken 7B Instruct v0.6 vs DeepSeek-R1-Distill-Qwen-7B (7.6B, index 39.8)
Sources
Something wrong here? Report it.
Run it locally
No GGUF build from a well-known publisher yet.
pip install -U mlx-lm
mlx_lm.chat --model mlx-community/Teuken-7B-instruct-v0.6-4bit-mlxvllm serve openGPT-X/Teuken-7B-instruct-v0.6Memory needed: weights, context cache and runtime buffers.
| Quantization | 4K context |
|---|---|
| 16-bit F16 | 16 GB |
| 8-bit Q8_0 | 8.7 GB |
| 4-bit Q4_K_M | 5.3 GB |