Client Portal Access
Real-time monitoring and management of your AI services
cognitus-h100-01:~$ nvidia-smi +-----------------------------------------------------------------------------+ | NVIDIA-SMI 535.129.03 Driver Version: 535.129.03 CUDA Version: 12.2 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | |===============================+======================+======================| | 0 NVIDIA H100 80GB On | 00000000:01:00.0 Off | 0 | | N/A 72C P0 650W / 700W | 67328MiB / 81920MiB | 67% Default | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 1337 C python3 llama_training.py 45123MiB | | 0 N/A N/A 2891 C python3 inference_server.py 22205MiB | +-----------------------------------------------------------------------------+
[Training] Epoch 3/5 | Step 9600/16000 [Training] Loss: 0.342 | Val Loss: 0.398 | Accuracy: 87.3% [Training] Tokens/sec: 45,230 | GPU Util: 98% [Training] Est. time remaining: 14h 22m [Training] Checkpoint saved: llama-3.1-70b-agrots-ep3-s9600.pt [Training] Memory allocated: 67.3GB / 80GB [Optimizer] Learning rate: 2.1e-5 [Dataset] Processed 2.3M / 2.3M samples [Validation] Running validation on 50k samples... [Validation] Accuracy: 87.3% | F1 Score: 0.891 [Training] Resuming training...
| Checkpoint Name | Epoch | Loss | Size | Action |
|---|---|---|---|---|
| llama-3.1-70b-agrots-ep3-s9600.pt | 3 | 0.342 | 142 GB | |
| llama-3.1-70b-agrots-ep2-s6400.pt | 2 | 0.621 | 142 GB | |
| llama-3.1-70b-agrots-ep1-s3200.pt | 1 | 0.987 | 142 GB |
# Using AWS CLI aws s3 cp s3://cognitus-models/llama-3.1-70b-agrots-ep3-s9600.pt ./ # Using curl curl -O https://models.cognituslab.com/checkpoints/llama-3.1-70b-agrots-ep3-s9600.pt # Verify checksum sha256sum llama-3.1-70b-agrots-ep3-s9600.pt
The AgriGPT API provides access to our specialized GPT model optimized for the agro-food sector. The API supports real-time inference, batch processing, and streaming responses.
curl -X POST https://api.cognituslab.com/v1/chat/completions \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json"
| Endpoint | Method | Description |
|---|---|---|
| /v1/chat/completions | POST | Generate chat completions |
| /v1/embeddings | POST | Generate embeddings |
| /v1/models | GET | List available models |
| /v1/batch | POST | Submit batch processing job |
{
"model": "agrigpt-v2.4",
"messages": [
{
"role": "system",
"content": "You are an expert in agro-food supply chain management."
},
{
"role": "user",
"content": "What are the optimal storage conditions for fresh tomatoes?"
}
],
"temperature": 0.7,
"max_tokens": 500
}
{
"id": "chatcmpl-8vQx2K4N",
"object": "chat.completion",
"model": "agrigpt-v2.4",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Optimal storage conditions for fresh tomatoes..."
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 45,
"completion_tokens": 127,
"total_tokens": 172
}
}