v0.1.0-rc0

Docker Images

• nilai-api: public.ecr.aws/k5d9x2g2/nilai-api:v0.1.0-rc0
• nilai-vllm: public.ecr.aws/k5d9x2g2/nilai-vllm:v0.1.0-rc0
• nilai-attestation: public.ecr.aws/k5d9x2g2/nilai-attestation:v0.1.0-rc0

What's Changed

• feat: added docker integration with Dockerfile + Compose by @jcabrero in #7
• feat: Add support for API keys by @jcabrero in #9
• feat: Added HTTPS and fixed various performance issues for CPU by @jcabrero in #10
• docs: Added OpenAPI and FastAPI docs in API docs page by @jcabrero in #13
• feat: Add query number by @jcabrero in #14
• feat: AMD SEV-SNP Attestation by @jcabrero in #18
• feat: Updated to dynamic model addition by @jcabrero in #25
• feat: Added Llama 8B option by @jcabrero in #26
• feat: Add Secret Models by @jcabrero in #27
• fix: openai api corrections by @jcabrero in #28
• feat: Add staggered docker compose deploy by @jcabrero in #29
• Feat: CI Pipelines by @jcabrero in #30
• feat: Allow streamed model responses by @jcabrero in #33
• fix: Streaming LLM and API outputs by @jcabrero in #35
• feat: Added maximum supported context windows for each model by @jcabrero in #36
• fix: Models disconnecting from etcd3 by @jcabrero in #39
• chore: Add better instructions for manual execution to README.md by @jcabrero in #42
• feat: add nilrag feature to nilai by @manel1874 in #43
• fix: updated numpy and nilrag versions by @jcabrero in #45
• feat: Add Llama 1B and 8B vLLM models for GPU by @mathias-atla in #40
• feat: new vLLM backend by @jcabrero in #46
• feat: Add cloud remote database for instances by @jcabrero in #48
• feat: Support for Nvidia CC by @jcabrero in #47
• fix: verifier gunicorn rollout by @jcabrero in #49
• chore: Add postgres docker compose volume by @jcabrero in #63
• feat: added dolphin models and changed logging to warning by @jcabrero in #65
• chore: Open Source Changes by @jcabrero in #66
• chore: fix pyright in nilai_api by @lumasepa in #68
• More pyright fixes by @lumasepa in #70
• fix: nilRAG GPU Usage on nilAI by @jcabrero in #71
• e2e tests by @lumasepa in #67
• feat: Add Llama 70B by @jcabrero in #76
• feat: DB Grafana Dashboards by @jcabrero in #91
• feat: add async to nilrag and nildb call by @manel1874 in #92
• fix: .env file clarifications and grafana access by @jcabrero in #95
• refactor: update nilrag operation to use new nilrag api by @manel1874 in #94
• Update README with macOS Local Execution Troubleshooting Instructions by @blefo in #69
• feat: Improved attestation verification by @jcabrero in #96
• feat: NUC authentication by @jcabrero in #98
• feat: Dual Authentication Mechanism API + NUC by @jcabrero in #101
• feat: new improved grafana dashboard and dashboard for nucs by @jcabrero in #102
• feat: allow all CORS origins by @jcabrero in #103
• feat: added docs token auth bypass by @jcabrero in #104
• feat: Update NUCs to compatibility with public keys by @jcabrero in #110
• chore: Update to rc version of nuc-py by @jcabrero in #112
• chore: add CD pipeline for nilAI images by @jcabrero in #113

Full Changelog: https://github.com/NillionNetwork/nilAI/commits/v0.1.0-rc0