Overview
Your compute savings and cluster health — May 2026.
Compute Savings
$2,457
vs. Anthropic Claude Sonnet 4.6
420M input tokens × $3.00/M = $1,260
140M output tokens × $15.00/M = $2,100
Anthropic equiv: $3,360 | You paid: $903
Also: $2,155 saved vs. GPT-4o
Energy Savings
$2,657
vs. ERCOT Houston commercial grid
70 kW × 730 hrs = 51,100 kWh consumed
Your rate: $0.030/kWh → Actual: $1,533
Houston ERCOT avg: $0.082/kWh → Grid equiv: $4,190
CO₂e Avoided
19.7 MT
metric tons avoided this month
51,100 kWh × 0.386 kg CO₂e/kWh ÷ 1,000
EPA eGRID 2023, Texas ERCT subregion
Equiv. to removing ~4.3 cars from the road for a year
Health Research Funded
$1,050
to American biomedical AI research
35% of your subscription above BotINFRA's operating threshold
This month: $1,050
Lifetime total: $8,925
Funds subsidized compute for open-source biomedical AI.
Tokens This Month
560.0M
API Calls
18,432
Actual Cost
$903
Cluster Status
Operational
24-Hour Activity
Updated every 5 min
0:003:006:009:0012:0015:0018:0021:00
Usage
Token consumption, spend, and energy for your account.
Daily Tokens — May 2026
May 1May 7May 14May 21May 28
By Model
| Model | Tokens | Calls | Actual | Anthropic Equiv. | Savings |
|---|---|---|---|---|---|
| Qwen2.5-14B-Instruct-AWQ | 448M | 14,746 | $734 | $3,136 | $2,402 |
| Qwen2.5-7B-Instruct-AWQ | 112M | 3,686 | $169 | $784 | $615 |
By Agent
| Agent ID | Tokens | Calls | Cost |
|---|---|---|---|
| agent_01JX4K2P… | 210M | 7,200 | $315 |
| agent_01JX3M8R… | 168M | 5,541 | $252 |
| agent_01JX1Q9T… | 112M | 3,686 | $168 |
| agent_01JWZN5V… | 70M | 2,005 | $108 |
Rolling 30-day window. Month-to-date totals on "This Month" align with calendar billing.
By Model — Last 30 Days
| Model | Tokens | Calls | Actual | Anthropic Equiv. | Savings |
|---|---|---|---|---|---|
| Qwen2.5-14B-Instruct-AWQ | 512M | 16,842 | $838 | $3,584 | $2,746 |
12-Month Summary
| Month | Tokens | Calls | Actual | Anthropic Equiv. | Savings |
|---|---|---|---|---|---|
| May 2026 | 560M | 18,432 | $903 | $3,360 | $2,457 |
| Apr 2026 | 498M | 16,104 | $813 | $2,934 | $2,121 |
| Mar 2026 | 445M | 14,218 | $728 | $2,620 | $1,892 |
| Feb 2026 | 380M | 12,100 | $622 | $2,238 | $1,616 |
| Jan 2026 | 320M | 10,240 | $524 | $1,884 | $1,360 |
kWh This Month
51,100
70 kW × 730 hrs
Your Rate
$0.030
per kWh — contracted Jan 2025
Grid Rate (ERCOT)
$0.082
Houston commercial avg
Rate Comparison
| Source | Rate / kWh | Monthly Cost | Savings vs You |
|---|---|---|---|
| Your Rate (ROS-01) | $0.030 | $1,533 | — |
| Houston ERCOT Avg | $0.082 | $4,190 | +$2,657 |
| Texas Commercial Avg | $0.091 | $4,650 | +$3,117 |
| US National Avg | $0.120 | $6,132 | +$4,599 |
Models
Deployed inference models and deployment requests.
Deployed Models
Qwen/Qwen2.5-14B-Instruct-AWQ
LIVE
Endpoint: http://176.9.158.22:8000/v1
Quantization: AWQ 4-bit
VRAM
Request Model Deployment
API Keys
Manage authentication keys for the BotINFRA API.
Your Keys
| Name | Key | Created | Last Used | Calls / Month | |
|---|---|---|---|---|---|
| Production | bi_live_•••••••• | Jan 14, 2025 | May 8, 2026 | 14,210 | |
| Staging | bi_live_•••••••• | Mar 3, 2026 | May 7, 2026 | 4,222 |
Create New Key
Cluster
Inference infrastructure status and SLA performance.
GPU Utilization
—
Monitoring coming soon
VRAM Used / Total
18.4 / 20 GB
RTX 4000 SFF Ada
Uptime This Month
—
Monitoring coming soon
P95 TTFT (1h)
—
Monitoring coming soon
Inference Plane — 176.9.158.22
Host
176.9.158.22
GPU
NVIDIA RTX 4000 SFF Ada
VRAM
20 GB
Runtime
vllm/vllm-openai:latest
API Port
8000 (OpenAI-compatible)
Model Loaded
Qwen/Qwen2.5-14B-Instruct-AWQ
Quantization
AWQ 4-bit
Model Cache
/data/models
Firewall
Port 8000 → 178.156.143.185 only
SLA — Time to First Token (P95)
| Model Class | P95 Commitment | Current | Status |
|---|---|---|---|
| 7B–14B | 500 ms | — | Monitoring pending |
| 30B+ | 1,200 ms | — | Monitoring pending |
TTFT instrumentation via vLLM metrics endpoint — on roadmap.