https://sonar.dphn.ai/https://sonar.dphn.ai/adapters/lora/https://sonar.dphn.ai/adapters/soft-prompts/https://sonar.dphn.ai/developer/adding-model/https://sonar.dphn.ai/developer/cutlass-epilogue/https://sonar.dphn.ai/developer/endpoint-plugins/https://sonar.dphn.ai/developer/input-processing/https://sonar.dphn.ai/developer/multimodal/https://sonar.dphn.ai/developer/paged-attention/https://sonar.dphn.ai/developer/tool-plugin/https://sonar.dphn.ai/developer/torch-custom-class/https://sonar.dphn.ai/installation/installation-cpu/https://sonar.dphn.ai/installation/installation-neuron/https://sonar.dphn.ai/installation/installation-rocm/https://sonar.dphn.ai/installation/installation-tpu/https://sonar.dphn.ai/installation/installation-windows/https://sonar.dphn.ai/installation/installation-xpu/https://sonar.dphn.ai/installation/installation/https://sonar.dphn.ai/prompt-caching/implementation/https://sonar.dphn.ai/prompt-caching/introduction/https://sonar.dphn.ai/quantization/kv-cache/https://sonar.dphn.ai/quantization/quantization-methods/https://sonar.dphn.ai/quantization/support-matrix/https://sonar.dphn.ai/spec-decoding/draft-model/https://sonar.dphn.ai/spec-decoding/mlpspeculator/https://sonar.dphn.ai/spec-decoding/ngram/https://sonar.dphn.ai/spec-decoding/overview/https://sonar.dphn.ai/usage/1-getting-started/https://sonar.dphn.ai/usage/debugging/https://sonar.dphn.ai/usage/derenderer/https://sonar.dphn.ai/usage/distributed/https://sonar.dphn.ai/usage/encoder-decoder/https://sonar.dphn.ai/usage/metrics/https://sonar.dphn.ai/usage/models/https://sonar.dphn.ai/usage/openai/https://sonar.dphn.ai/usage/optimization/https://sonar.dphn.ai/usage/per-request-metrics/https://sonar.dphn.ai/usage/tool-call/https://sonar.dphn.ai/usage/vlm/