Skip to content

Commit b3b6c53

Browse files
T-DevHT-DevH
committed
fix: correct markdown formatting in Step 4 notebook cell
- Fix line breaks and formatting in Step 4 markdown cell - Ensure proper newline handling in notebook JSON structure - Improve markdown cell structure for better rendering
1 parent dc3602f commit b3b6c53

File tree

1 file changed

+92
-82
lines changed

1 file changed

+92
-82
lines changed

‎notebooks/setup/complete_setup_guide.ipynb‎

Lines changed: 92 additions & 82 deletions
Original file line numberDiff line numberDiff line change
@@ -464,91 +464,101 @@
464464
"cell_type": "markdown",
465465
"metadata": {},
466466
"source": [
467-
"## Step 4: API Key Configuration (NVIDIA & Brev)",
468-
"",
469-
"The Warehouse Operational Assistant uses NVIDIA NIMs (NVIDIA Inference Microservices) for AI capabilities. You have **two deployment options** for NIMs:",
470-
"",
471-
"### \ud83d\ude80 NIM Deployment Options",
472-
"",
473-
"**Option 1: Cloud Endpoints** (Easiest - Default)",
474-
"- Use NVIDIA's cloud-hosted NIM services",
475-
"- **No installation required** - just configure API keys",
476-
"- Quick setup, perfect for development and testing",
477-
"- Endpoints: `api.brev.dev` or `integrate.api.nvidia.com`",
478-
"",
479-
"**Option 2: Self-Hosted NIMs** (Recommended for Production)",
480-
"- **Install NIMs on your own infrastructure** using Docker",
481-
"- **Create custom endpoints** on your servers",
482-
"- Benefits:",
483-
" - \ud83d\udd12 **Data Privacy**: Keep sensitive data on-premises",
484-
" - \ud83d\udcb0 **Cost Control**: Avoid per-request cloud costs",
485-
" - \u2699\ufe0f **Custom Requirements**: Full control over infrastructure",
486-
" - \u26a1 **Low Latency**: Reduced network latency",
487-
"",
488-
"**Self-Hosting Example:**",
489-
"```bash",
490-
"# Deploy LLM NIM on your server",
491-
"docker run --gpus all -p 8000:8000 \\",
492-
" nvcr.io/nvidia/nim/llama-3.3-nemotron-super-49b:latest",
493-
"",
494-
"# Then set in .env:",
495-
"LLM_NIM_URL=http://your-server:8000/v1",
496-
"```",
497-
"",
498-
"**\ud83d\udcdd Note**: This step configures API keys for cloud endpoints. If you're self-hosting NIMs, you can skip API keys (unless your NIMs require authentication) and just configure the endpoint URLs in Step 5.",
499-
"",
500-
"---",
501-
"",
502-
"### \u26a0\ufe0f Important: Two Types of API Keys (for Cloud Endpoints)",
503-
"",
504-
"**1. NVIDIA API Key** (starts with `nvapi-`)",
505-
"- **Format**: `nvapi-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx`",
506-
"- **Get from**: https://build.nvidia.com/",
507-
"- **Works with**: Both `api.brev.dev` and `integrate.api.nvidia.com` endpoints",
508-
"- **Required for**: Embedding service (always requires NVIDIA API key)",
509-
"",
510-
"**2. Brev API Key** (starts with `brev_api_`)",
511-
"- **Format**: `brev_api_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx`",
512-
"- **Get from**: Your Brev account",
513-
"- **Works with**: `api.brev.dev` endpoint only",
514-
"- **Optional**: Can use NVIDIA API key instead",
515-
"",
516-
"### Configuration Options (Cloud Endpoints)",
517-
"",
518-
"**Option A: Use NVIDIA API Key for Everything** (Recommended)",
519-
"- Set `NVIDIA_API_KEY` with your NVIDIA API key",
520-
"- Leave `EMBEDDING_API_KEY` unset (will use `NVIDIA_API_KEY`)",
521-
"- Works with both endpoints",
522-
"",
523-
"**Option B: Use Brev API Key for LLM + NVIDIA API Key for Embedding**",
524-
"- Set `NVIDIA_API_KEY` with your Brev API key",
525-
"- **MUST** set `EMBEDDING_API_KEY` with your NVIDIA API key (required!)",
526-
"- Embedding service always requires NVIDIA API key",
527-
"",
528-
"### Getting Your API Keys (for Cloud Endpoints)",
529-
"",
530-
"**NVIDIA API Key:**",
531-
"1. **Visit**: https://build.nvidia.com/",
532-
"2. **Sign up** or log in to your NVIDIA account",
533-
"3. **Navigate** to the \"API Keys\" section",
534-
"4. **Create** a new API key",
535-
"5. **Copy** the API key (starts with `nvapi-`)",
536-
"",
537-
"**Brev API Key (Optional):**",
538-
"1. **Visit**: Your Brev account dashboard",
539-
"2. **Navigate** to API Keys section",
540-
"3. **Create** or copy your Brev API key (starts with `brev_api_`)",
541-
"",
542-
"### What You'll Get Access To",
543-
"",
544-
"- **LLM Service** (Llama 3.3 Nemotron Super 49B) - for chat and reasoning",
545-
"- **Embedding Service** (llama-3_2-nv-embedqa-1b-v2) - for semantic search",
546-
"- **Document Processing** - OCR and structured data extraction",
547-
"- **Content Safety** - NeMo Guardrails for content moderation",
548-
"",
467+
"## Step 4: API Key Configuration (NVIDIA & Brev)\n",
468+
"\n",
469+
"The Warehouse Operational Assistant uses NVIDIA NIMs (NVIDIA Inference Microservices) for AI capabilities. You have **two deployment options** for NIMs:\n",
470+
"\n",
471+
"### \ud83d\ude80 NIM Deployment Options\n",
472+
"\n",
473+
"**Option 1: Cloud Endpoints** (Easiest - Default)\n",
474+
"- Use NVIDIA's cloud-hosted NIM services\n",
475+
"- **No installation required** - just configure API keys\n",
476+
"- Quick setup, perfect for development and testing\n",
477+
"- Endpoints: `api.brev.dev` or `integrate.api.nvidia.com`\n",
478+
"\n",
479+
"**Option 2: Self-Hosted NIMs** (Recommended for Production)\n",
480+
"- **Install NIMs on your own infrastructure** using Docker\n",
481+
"- **Create custom endpoints** on your servers\n",
482+
"- Benefits:\n",
483+
" - \ud83d\udd12 **Data Privacy**: Keep sensitive data on-premises\n",
484+
" - \ud83d\udcb0 **Cost Control**: Avoid per-request cloud costs\n",
485+
" - \u2699\ufe0f **Custom Requirements**: Full control over infrastructure\n",
486+
" - \u26a1 **Low Latency**: Reduced network latency\n",
487+
"\n",
488+
"**Self-Hosting Example:**\n",
489+
"```bash\n",
490+
"# Deploy LLM NIM on your server\n",
491+
"docker run --gpus all -p 8000:8000 \\\n",
492+
" nvcr.io/nvidia/nim/llama-3.3-nemotron-super-49b:latest\n",
493+
"\n",
494+
"# Then set in .env:\n",
495+
"LLM_NIM_URL=http://your-server:8000/v1\n",
496+
"```\n",
497+
"\n",
498+
"**\ud83d\udcdd Note**: This step configures API keys for cloud endpoints. If you're self-hosting NIMs, you can skip API keys (unless your NIMs require authentication) and just configure the endpoint URLs in Step 5.\n",
499+
"\n",
500+
"---\n",
501+
"\n",
502+
"### \u26a0\ufe0f Important: Two Types of API Keys (for Cloud Endpoints)\n",
503+
"\n",
504+
"**1. NVIDIA API Key** (starts with `nvapi-`)\n",
505+
"- **Format**: `nvapi-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx`\n",
506+
"- **Get from**: https://build.nvidia.com/\n",
507+
"- **Works with**: Both `api.brev.dev` and `integrate.api.nvidia.com` endpoints\n",
508+
"- **Required for**: Embedding service (always requires NVIDIA API key)\n",
509+
"\n",
510+
"**2. Brev API Key** (starts with `brev_api_`)\n",
511+
"- **Format**: `brev_api_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx`\n",
512+
"- **Get from**: Your Brev account\n",
513+
"- **Works with**: `api.brev.dev` endpoint only\n",
514+
"- **Optional**: Can use NVIDIA API key instead\n",
515+
"\n",
516+
"### Configuration Options (Cloud Endpoints)\n",
517+
"\n",
518+
"**Option A: Use NVIDIA API Key for Everything** (Recommended)\n",
519+
"- Set `NVIDIA_API_KEY` with your NVIDIA API key\n",
520+
"- Leave `EMBEDDING_API_KEY` unset (will use `NVIDIA_API_KEY`)\n",
521+
"- Works with both endpoints\n",
522+
"\n",
523+
"**Option B: Use Brev API Key for LLM + NVIDIA API Key for Embedding**\n",
524+
"- Set `NVIDIA_API_KEY` with your Brev API key\n",
525+
"- **MUST** set `EMBEDDING_API_KEY` with your NVIDIA API key (required!)\n",
526+
"- Embedding service always requires NVIDIA API key\n",
527+
"\n",
528+
"### Getting Your API Keys (for Cloud Endpoints)\n",
529+
"\n",
530+
"**NVIDIA API Key:**\n",
531+
"1. **Visit**: https://build.nvidia.com/\n",
532+
"2. **Sign up** or log in to your NVIDIA account\n",
533+
"3. **Navigate** to the \"API Keys\" section\n",
534+
"4. **Create** a new API key\n",
535+
"5. **Copy** the API key (starts with `nvapi-`)\n",
536+
"\n",
537+
"**Brev API Key (Optional):**\n",
538+
"1. **Visit**: Your Brev account dashboard\n",
539+
"2. **Navigate** to API Keys section\n",
540+
"3. **Create** or copy your Brev API key (starts with `brev_api_`)\n",
541+
"\n",
542+
"### What You'll Get Access To\n",
543+
"\n",
544+
"- **LLM Service** (Llama 3.3 Nemotron Super 49B) - for chat and reasoning\n",
545+
"- **Embedding Service** (llama-3_2-nv-embedqa-1b-v2) - for semantic search\n",
546+
"- **Document Processing** - OCR and structured data extraction\n",
547+
"- **Content Safety** - NeMo Guardrails for content moderation\n",
548+
"\n",
549549
"**\ud83d\udca1 For Self-Hosted NIMs**: See `DEPLOYMENT.md` section \"NVIDIA NIMs Deployment & Configuration\" for detailed self-hosting instructions."
550550
]
551551
},
552+
{
553+
"cell_type": "markdown",
554+
"metadata": {},
555+
"source": []
556+
},
557+
{
558+
"cell_type": "markdown",
559+
"metadata": {},
560+
"source": []
561+
},
552562
{
553563
"cell_type": "code",
554564
"execution_count": null,

0 commit comments

Comments
 (0)