feat: add all OCI GenAI models — 26 chat models and 9 embedding models

Complete model catalog from Oracle Cloud Infrastructure GenAI docs:
- Cohere: Command A Reasoning, Vision, A, R+, R
- Meta: Llama 4 Maverick/Scout, 3.3 70B, 3.2 90B/11B Vision, 3.1 405B
- Google: Gemini 2.5 Pro, Flash, Flash-Lite
- xAI: Grok Code Fast 1, 4.1 Fast, 4 Fast, 4, 3, 3 Mini, 3 Fast, 3 Mini Fast
- OpenAI: GPT-oss 120B, 20B
- Embeddings: Cohere Embed v4.0, English/Multilingual v3.0 + Light + Image variants
- Updated model IDs to match official OCI SDK format
This commit is contained in:
nogueiraguh
2026-03-03 19:07:38 -03:00
parent 09371cb675
commit 723e2a47a1
2 changed files with 40 additions and 7 deletions

View File

@@ -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

View File

@@ -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 = [