Dashboard

NovAI Labs · Tenant View

8 GPUs Active May 2026

Welcome back, NovAI Labs!

Your GPU compute workspace. 8 GPUs allocated, 3 jobs running, $18,420 consumed this period.

Enterprise Plan 100 GPU Quota us-east-1 us-west-2
Allocated GPUs
8
of 100 quota used
↑ 2 added this period
Active Jobs
3
5 completed · 2 queued
↑ 1 from yesterday
Period Cost
$18.4K
May 1–18, 2026
72% of budget
GPU-Hours
2,847
consumed this month
↓ 12% vs last month
Recent Activity 10
LLaMA-3 Fine-tune running
4× H100 · Job #J-0041 · Epoch 7/20
2m ago
Stable Diffusion XL Training running
2× A100 · Job #J-0040 · 34% complete
15m ago
Embedding Batch Inference running
1× A100 · Job #J-0039 · 68% complete
28m ago
Whisper-v3 Transcription queued
1× A100 · Job #J-0042 · Waiting for GPU
5m ago
GPT-4 Distillation Run done
4× H100 · Job #J-0038 · 6h 14m total
3h ago
Mistral-7B Evaluation done
2× A100 · Job #J-0037 · 1h 42m total
7h ago
Allocated GPUs 8 active
Quota: 8 / 100 used
H100 SXM5
80 GB HBM3
BUSY
gpu-h100-01 us-east-1a node-04
GPU Utilization94%
76°C
Temp
620W
Power
3.9 TB/s
Mem BW
Job: LLaMA-3 Fine-tune · #J-0041
H100 SXM5
80 GB HBM3
BUSY
gpu-h100-02 us-east-1a node-04
GPU Utilization97%
79°C
Temp
635W
Power
3.9 TB/s
Mem BW
Job: LLaMA-3 Fine-tune · #J-0041
H100 SXM5
80 GB HBM3
BUSY
gpu-h100-03 us-east-1b node-05
GPU Utilization91%
74°C
Temp
608W
Power
3.8 TB/s
Mem BW
Job: LLaMA-3 Fine-tune · #J-0041
H100 SXM5
80 GB HBM3
BUSY
gpu-h100-04 us-east-1b node-05
GPU Utilization88%
71°C
Temp
594W
Power
3.7 TB/s
Mem BW
Job: LLaMA-3 Fine-tune · #J-0041
A100 PCIe
80 GB HBM2e
BUSY
gpu-a100-01 us-west-2a node-12
GPU Utilization73%
62°C
Temp
310W
Power
2.0 TB/s
Mem BW
Job: SD-XL Training · #J-0040
A100 PCIe
80 GB HBM2e
BUSY
gpu-a100-02 us-west-2a node-12
GPU Utilization69%
58°C
Temp
292W
Power
1.9 TB/s
Mem BW
Job: SD-XL Training · #J-0040
A100 PCIe
80 GB HBM2e
BUSY
gpu-a100-03 us-west-2b node-13
GPU Utilization82%
66°C
Temp
336W
Power
2.1 TB/s
Mem BW
Job: Embedding Inference · #J-0039
A100 PCIe
80 GB HBM2e
IDLE
gpu-a100-04 us-west-2b node-13
GPU Utilization4%
38°C
Temp
48W
Power
Mem BW
No active job · queued: Whisper-v3
Job History 10 jobs
3 running · 2 queued · 5 done
Job ID Name / Type GPUs Status Progress Duration Cost
#J-0041
LLaMA-3 Fine-tune
Training · High priority
4× H100 running
35%
2h 14m $1,246
#J-0040
SD-XL Training
Training · Normal priority
2× A100 running
34%
1h 52m $386
#J-0039
Embedding Batch Inference
Inference · Normal priority
1× A100 running
68%
0h 44m $88
#J-0042
Whisper-v3 Transcription
Inference · Normal priority
1× A100 queued
0%
#J-0043
RLHF Reward Model
Training · High priority
2× H100 queued
0%
#J-0038
GPT-4 Distillation
Training · High priority
4× H100 done
100%
6h 14m $3,492
#J-0037
Mistral-7B Evaluation
Inference · Normal priority
2× A100 done
100%
1h 42m $354
#J-0036
CLIP Vision Encoder
Training · Normal priority
2× A100 done
100%
4h 22m $906
#J-0035
DPO Alignment Run
Training · Normal priority
2× H100 done
100%
3h 08m $2,180
#J-0034
Tokenizer Benchmark
Evaluation · Low priority
1× A100 done
100%
0h 38m $76
Current Period Spend
$18,420
May 1–18, 2026 · of $25,500 budget
GPU-Hours Consumed
2,847h
Across all GPU types
Effective Rate
$6.47/h
Blended avg across H100 + A100
Cost Breakdown by GPU Type
H100 SXM5
$11,422 · 1,142h
A100 PCIe
$6,840 · 1,705h
Other
$158 · 0h
Invoice History 5 invoices
Invoice Period GPU-Hours Amount Status Payment
INV-0018 May 1–18, 2026 2,847h $18,420 Pending View
INV-0017 Apr 1–30, 2026 3,241h $21,040 Paid Receipt
INV-0016 Mar 1–31, 2026 2,988h $19,360 Paid Receipt
INV-0015 Feb 1–28, 2026 2,540h $16,180 Paid Receipt
INV-0014 Jan 1–31, 2026 1,880h $11,800 Overdue Pay Now

API Keys

Manage authentication keys for programmatic GPU access. Keys are hashed on creation — copy on first view.

Production Workloads
gpf_prod_••••••••••••••••4a2f
read write submit_jobs
Created Mar 12, 2026
Last used 2m ago
CI/CD Pipeline
gpf_ci_••••••••••••••••9b1c
read submit_jobs
Created Apr 2, 2026
Last used 1h ago
Monitoring / Read-only
gpf_ro_••••••••••••••••3e7d
read
Created Jan 20, 2026
Last used 3d ago
Using the API
Pass your key as Authorization: Bearer <key> on all API requests. Keys are scoped to your tenant — you cannot query or modify other tenants' resources. See API Documentation →