diff --git a/README.md b/README.md index 8f3f746..7e33e03 100644 --- a/README.md +++ b/README.md @@ -32,7 +32,7 @@ The platform combines security compliance scanning, AI-powered chat with **RAG ( ### 🤖 AI Chat Agent with RAG - **OCI Generative AI** integration via official SDK (`oci.generative_ai_inference`) - **RAG (Retrieval-Augmented Generation)**: automatically queries ADB vector store for relevant context before generating responses -- 14 models across 5 providers: **Cohere** (Command A/R/R+), **Meta** (Llama 4/3.3/3.2/3.1), **Google** (Gemini 2.5), **xAI** (Grok 3/4), **OpenAI** (GPT-oss 120B/20B) +- 26 models across 5 providers: **Cohere** (Command A Reasoning/Vision/R/R+), **Meta** (Llama 4/3.3/3.2/3.1), **Google** (Gemini 2.5 Pro/Flash/Flash-Lite), **xAI** (Grok 4.1/4/3/3 Mini + fast variants), **OpenAI** (GPT-oss 120B/20B) - 16 OCI regions supported with auto-generated endpoints - Full parameter control: temperature, max_tokens, top_p, top_k, frequency/presence penalty - Conversation history with session management @@ -403,18 +403,28 @@ oci-cis-agent/ | Provider | Model | API Format | |----------|-------|------------| +| Cohere | Command A Reasoning | COHERE | +| Cohere | Command A Vision | COHERE | | Cohere | Command A (03-2025) | COHERE | | Cohere | Command R+ (08-2024) | COHERE | | Cohere | Command R (08-2024) | COHERE | | Meta | Llama 4 Maverick | GENERIC | | Meta | Llama 4 Scout | GENERIC | -| Meta | Llama 3.3 70B Instruct | GENERIC | +| Meta | Llama 3.3 (70B) | GENERIC | | Meta | Llama 3.2 90B Vision | GENERIC | -| Meta | Llama 3.1 405B Instruct | GENERIC | +| Meta | Llama 3.2 11B Vision | GENERIC | +| Meta | Llama 3.1 (405B) | GENERIC | | Google | Gemini 2.5 Pro | GENERIC | | Google | Gemini 2.5 Flash | GENERIC | +| Google | Gemini 2.5 Flash-Lite | GENERIC | +| xAI | Grok Code Fast 1 | GENERIC | +| xAI | Grok 4.1 Fast | GENERIC | +| xAI | Grok 4 Fast | GENERIC | | xAI | Grok 4 | GENERIC | | xAI | Grok 3 | GENERIC | +| xAI | Grok 3 Mini | GENERIC | +| xAI | Grok 3 Fast | GENERIC | +| xAI | Grok 3 Mini Fast | GENERIC | | OpenAI | GPT-oss (120B) | GENERIC | | OpenAI | GPT-oss (20B) | GENERIC | @@ -422,11 +432,15 @@ oci-cis-agent/ | Provider | Model | Dimensions | |----------|-------|------------| +| Cohere | Embed v4.0 (Multimodal) | 1536 | | Cohere | Embed English v3.0 | 1024 | | Cohere | Embed Multilingual v3.0 | 1024 | | Cohere | Embed English Light v3.0 | 384 | | Cohere | Embed Multilingual Light v3.0 | 384 | -| Cohere | Embed v4.0 (Multimodal) | 1536 | +| Cohere | Embed English Image v3.0 | 1024 | +| Cohere | Embed Multilingual Image v3.0 | 1024 | +| Cohere | Embed English Light Image v3.0 | 384 | +| Cohere | Embed Multilingual Light Image v3.0 | 384 | ### GenAI Regions diff --git a/backend/app.py b/backend/app.py index 534f582..8ad1b22 100644 --- a/backend/app.py +++ b/backend/app.py @@ -46,28 +46,47 @@ security = HTTPBearer() # ── OCI GenAI Models Catalog ────────────────────────────────────────────────── GENAI_MODELS = { + # Cohere + "cohere.command-a-reasoning": {"provider":"cohere","name":"Cohere Command A Reasoning","api_format":"COHERE"}, + "cohere.command-a-vision": {"provider":"cohere","name":"Cohere Command A Vision","api_format":"COHERE"}, "cohere.command-a-03-2025": {"provider":"cohere","name":"Cohere Command A","api_format":"COHERE"}, "cohere.command-r-plus-08-2024": {"provider":"cohere","name":"Cohere Command R+ (08-2024)","api_format":"COHERE"}, "cohere.command-r-08-2024": {"provider":"cohere","name":"Cohere Command R (08-2024)","api_format":"COHERE"}, - "meta.llama-4-maverick": {"provider":"meta","name":"Meta Llama 4 Maverick","api_format":"GENERIC"}, - "meta.llama-4-scout": {"provider":"meta","name":"Meta Llama 4 Scout","api_format":"GENERIC"}, + # Meta + "meta.llama-4-maverick-17b-128e-instruct-fp8": {"provider":"meta","name":"Meta Llama 4 Maverick","api_format":"GENERIC"}, + "meta.llama-4-scout-17b-16e-instruct": {"provider":"meta","name":"Meta Llama 4 Scout","api_format":"GENERIC"}, "meta.llama-3.3-70b-instruct": {"provider":"meta","name":"Meta Llama 3.3 (70B)","api_format":"GENERIC"}, "meta.llama-3.2-90b-vision-instruct": {"provider":"meta","name":"Meta Llama 3.2 90B Vision","api_format":"GENERIC"}, + "meta.llama-3.2-11b-vision-instruct": {"provider":"meta","name":"Meta Llama 3.2 11B Vision","api_format":"GENERIC"}, "meta.llama-3.1-405b-instruct": {"provider":"meta","name":"Meta Llama 3.1 (405B)","api_format":"GENERIC"}, + # Google "google.gemini-2.5-pro": {"provider":"google","name":"Google Gemini 2.5 Pro","api_format":"GENERIC"}, "google.gemini-2.5-flash": {"provider":"google","name":"Google Gemini 2.5 Flash","api_format":"GENERIC"}, + "google.gemini-2.5-flash-lite": {"provider":"google","name":"Google Gemini 2.5 Flash-Lite","api_format":"GENERIC"}, + # xAI + "xai.grok-code-fast-1": {"provider":"xai","name":"xAI Grok Code Fast 1","api_format":"GENERIC"}, + "xai.grok-4.1-fast": {"provider":"xai","name":"xAI Grok 4.1 Fast","api_format":"GENERIC"}, + "xai.grok-4-fast": {"provider":"xai","name":"xAI Grok 4 Fast","api_format":"GENERIC"}, "xai.grok-4": {"provider":"xai","name":"xAI Grok 4","api_format":"GENERIC"}, "xai.grok-3": {"provider":"xai","name":"xAI Grok 3","api_format":"GENERIC"}, + "xai.grok-3-mini": {"provider":"xai","name":"xAI Grok 3 Mini","api_format":"GENERIC"}, + "xai.grok-3-fast": {"provider":"xai","name":"xAI Grok 3 Fast","api_format":"GENERIC"}, + "xai.grok-3-mini-fast": {"provider":"xai","name":"xAI Grok 3 Mini Fast","api_format":"GENERIC"}, + # OpenAI "openai.gpt-oss-120b": {"provider":"openai","name":"OpenAI GPT-oss (120B)","api_format":"GENERIC"}, "openai.gpt-oss-20b": {"provider":"openai","name":"OpenAI GPT-oss (20B)","api_format":"GENERIC"}, } EMBEDDING_MODELS = { + "cohere.embed-v4.0": {"name":"Cohere Embed v4.0 (Multimodal)","dims":1536}, "cohere.embed-english-v3.0": {"name":"Cohere Embed English v3.0","dims":1024}, "cohere.embed-multilingual-v3.0": {"name":"Cohere Embed Multilingual v3.0","dims":1024}, "cohere.embed-english-light-v3.0": {"name":"Cohere Embed English Light v3.0","dims":384}, "cohere.embed-multilingual-light-v3.0": {"name":"Cohere Embed Multilingual Light v3.0","dims":384}, - "cohere.embed-v4.0": {"name":"Cohere Embed v4.0 (Multimodal)","dims":1536}, + "cohere.embed-english-image-v3.0": {"name":"Cohere Embed English Image v3.0","dims":1024}, + "cohere.embed-multilingual-image-v3.0": {"name":"Cohere Embed Multilingual Image v3.0","dims":1024}, + "cohere.embed-english-light-image-v3.0": {"name":"Cohere Embed English Light Image v3.0","dims":384}, + "cohere.embed-multilingual-light-image-v3.0": {"name":"Cohere Embed Multilingual Light Image v3.0","dims":384}, } GENAI_REGIONS = [