Swiss AI
Apertus v1.5 8B
Atual
602Downloads23CurtidasJul 2026Publicado262K tokensContextoApache 2.0Licença58 BomQualidade
Apertus v1.5 8B (8.899999618530273B parameters) requires approximately 8.9 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 11 GB of VRAM.
Comece agora
— copie e cole para rodar localmenteCopy-paste commands to run Apertus v1.5 8B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "swiss-ai/Apertus-v1.5-8B" \
--hf-file "Apertus-v1.5-8B-Q4_K_M.gguf" \
-c 4096 -ngl 99Quick specs
Parameters8.9B
Architecturedense
Context262K tokens
Modalitytext+vision
Min RAM3.5 GB
Rec. RAM5.4 GB (Q4_K_M)
LicenseApache 2.0
FamilyApertus
✓ Chat
About this model
- •8.9B dense including a ~294M vision encoder — runs on a single consumer GPU at Q4.
- •Fully open training data, weights and documentation.
- •262,144-token context window.
- •19T pretraining tokens plus a 4T-token multimodal continued-pretraining mix.
- •Apache 2.0, developed by the Swiss AI initiative (EPFL / ETH Zurich).
Modelos relacionados
Escolhas rápidas
Melhor hardware
Melhores opções para Apertus v1.5 8B
Rodar este modelo
Opções de quantização
Estimativas de VRAM por nível de quantização
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q1_0_G128 | 1.125 | 1.3 GB | Very Low | — |
Q2_0_G128 | 1.71 | 2.4 GB | Low | — |
Q2_K | 2 | 3.5 GB | Low | — |
Q3_K_S | 3 | 4.4 GB | Low | — |
NVFP4 | 4 | 5.0 GB | Medium | — |
Q4_K_M | 4 | 5.4 GB | Medium | — |
Q5_K_M | 5 | 6.4 GB | High | — |
Q6_K | 6 | 7.3 GB | High | — |
Q8_0 | 8 | 9.5 GB | Very High | — |
F16 | 16 | 18.2 GB | Maximum | — |
Compatibilidade de hardware
Estimativas de compatibilidade para todo o hardware
Computing compatibility...
Detalhamento de memória
Reference: RTX 2060 6GB
Weights5.4 GB
KV Cache2.0 GB
Runtime0.9 GB
Headroom0.6 GB
Perguntas frequentes
FAQ — Apertus v1.5 8B
Veja também