Leaderboard / OpenLLM-France (Luciole)
Luciole 8B Instruct
OpenLLM-France (Luciole) · Luciole · released 2026-07-07 · data checked 2026-11-01
Luciole 8B Instruct is a 8.1B-parameter dense model from OpenLLM-France (Luciole) (Europe), released 2026-07-07 under the Apache-2.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 5.7 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.
No benchmark results for this release yet. Know of one? Send it.
Specifications
- Lab
- OpenLLM-France (Luciole) (FR · Europe)
- Architecture
- Dense transformer
- Total parameters
- 8.1B
- Active parameters
- All (dense)
- Size class
- Small
- Context window
- 131K tokens
- Input modalities
- text
- Reasoning
- Instruct (no thinking)
- License
- Apache-2.0
- Openness
- Open weights
- Memory to run
- ~5.7 GB at 4-bit, 8K context (more sizes)
- Hugging Face
- 3.9K downloads (30 days) · 12 likes
Compare with
- Luciole 8B Instruct vs Granite 4.2 8B (8B, index 63.5)
- Luciole 8B Instruct vs Ministral 3 8B (8B, index 55.2)
- Luciole 8B Instruct vs Gemma 4 E4B (8B, index 38.2)
Sources
Something wrong here? Report it.
Run it locally
Uses OpenLLM-France/Luciole-8B-Instruct-1.1-GGUF (Q4_K_M). Swap the tag for another size: BF16, Q4_K_M, Q5_K_M, Q8_0.
ollama run hf.co/OpenLLM-France/Luciole-8B-Instruct-1.1-GGUF:Q4_K_Mllama-server -hf OpenLLM-France/Luciole-8B-Instruct-1.1-GGUF:Q4_K_Mvllm serve OpenLLM-France/Luciole-8B-Instruct-1.1Memory needed: weights, context cache and runtime buffers.
| Quantization | 4K context | 32K context | 131K context |
|---|---|---|---|
| 16-bit F16 | 17 GB | 17 GB | 19 GB |
| 8-bit Q8_0 | 9.3 GB | 9.8 GB | 11 GB |
| 4-bit Q4_K_M | 5.6 GB | 6.1 GB | 7.7 GB |