Model Usage — per-model token analytics.
7-day token consumption, estimated cost, and daily request trends across all 7 gateway models. Costs use May 2026 list prices.
Est. Cost (7d)
$2.0645
based on list prices
Total Requests
3,461
across all models
Total Tokens
2946.9K
input + output
Per-model breakdown
| Model | Requests | Input (K) | Output (K) | Est. Cost | % of Requests |
|---|
| Cerebras · Llama 70B | 1,461 | 996.0K | 266.5K | $0.7575 | |
| DeepSeek Chat | 897 | 611.0K | 152.8K | $0.3331 | |
| Groq · Llama 70B | 529 | 354.0K | 96.0K | $0.3246 | |
| Claude Haiku 4.5 | 354 | 213.0K | 64.0K | $0.4264 | |
| Gemini 2.5 Flash | 122 | 76.5K | 22.2K | $0.0476 | |
| DeepSeek Reasoner | 70 | 46.3K | 18.5K | $0.0660 | |
| Gemini 2.5 Pro | 28 | 21.9K | 8.2K | $0.1094 | |
Daily request trends
| Date | Cerebras · Llama 70B | Groq · Llama 70B | DeepSeek Chat | DeepSeek Reasoner | Claude Haiku 4.5 | Gemini 2.5 Pro | Gemini 2.5 Flash | Total |
|---|
| Jul 23 | 201 | 72 | 122 | — | 48 | — | — | 443 |
| Jul 24 | 218 | 81 | 135 | 17 | 53 | — | — | 504 |
| Jul 25 | 174 | 60 | 105 | — | 40 | 8 | 25 | 412 |
| Jul 26 | 235 | 88 | 148 | 20 | 60 | — | 35 | 586 |
| Jul 27 | 195 | 66 | 115 | 15 | 44 | 9 | — | 444 |
| Jul 28 | 228 | 84 | 142 | — | 57 | 11 | 33 | 555 |
| Jul 29 | 210 | 78 | 130 | 18 | 52 | — | 29 | 517 |
Cost breakdown
1Cerebras · Llama 70B$0.757536.7% 2Claude Haiku 4.5$0.426420.7% 3DeepSeek Chat$0.333116.1% 4Groq · Llama 70B$0.324615.7% 5Gemini 2.5 Pro$0.10945.3% 6DeepSeek Reasoner$0.06603.2% 7Gemini 2.5 Flash$0.04762.3%