Models Library
Browse the models pre-loaded on Regolo, ready to use and callable through APIs compatible with the OpenAI standard. All chat models can also be consumed via one of our subscriptions.
Didn't find the model you need?
Bring your custom model
Pick any vLLM-compatible model from Hugging Face, including your own private models, and deploy it on our GPUs in minutes. Host it with the same low latency, EU compliance, and OpenAI-compatible API as every model above.
Our inferece at your disposal
See custom model pricing
INDEPENDENT BENCHMARKS & MODEL GUIDES
Compare open-weight models by speed, cost per task, and coding accuracy
Browse our full model comparisons directory and European sovereign AI alternatives hub, or read our deep-dive evaluations:
Qwen3.8-27B Benchmarks →
Every test where 27B beats Claude Opus 4.6 Max DeepSeek V4 Flash vs Qwen3.8 vs GLM-5.3 →
Quality-to-price leader across Flash models DeepSeek 1017 vs Muse Spark 816B vs Beam 501B →
DeepSWE v1.1 vs SWE-bench Verified & cost per task Jev & System-1 Open-Weight Models →
Fast non-thinking models vs reasoning verbosity Bonsai 27B & Qwen3.8-27B Deployment →
VRAM footprint, quantization, and EU cloud hosting DeepSeek-OCR vs GLM-OCR vs PaddleOCR →
Document extraction accuracy and cost benchmarks Gemma 4 31B vs Qwen3.6-35B-A3B →
Dense 31B multimodal vs sparse MoE throughput MiniMax M3 vs GLM-5.2 →
When to use each frontier open-weight giant
Every test where 27B beats Claude Opus 4.6 Max DeepSeek V4 Flash vs Qwen3.8 vs GLM-5.3 →
Quality-to-price leader across Flash models DeepSeek 1017 vs Muse Spark 816B vs Beam 501B →
DeepSWE v1.1 vs SWE-bench Verified & cost per task Jev & System-1 Open-Weight Models →
Fast non-thinking models vs reasoning verbosity Bonsai 27B & Qwen3.8-27B Deployment →
VRAM footprint, quantization, and EU cloud hosting DeepSeek-OCR vs GLM-OCR vs PaddleOCR →
Document extraction accuracy and cost benchmarks Gemma 4 31B vs Qwen3.6-35B-A3B →
Dense 31B multimodal vs sparse MoE throughput MiniMax M3 vs GLM-5.2 →
When to use each frontier open-weight giant