| Model | Endpoint | Context | Input /M | Output /M | Cached input /M | Cache write /M | Uptime 1d |
|---|---|---|---|---|---|---|---|
 GLM 5.3 | akashml/fp4 | 1M | $0.19 | $4.40 | $0.19 | — | 100.00% |
 Gemma 4 26B A4B | akashml/fp4 | 131K | $0.07 | $0.24 | $0.046 | — | 99.41% |
 Kimi K3.1 Pro | akashml/fp4 | 1M | $1.75 | $10.00 | $0.18 | — | 99.72% |
 Llama 3.3 70B Instruct | akashml/fp8 | 131K | $0.20 | $0.52 | $0.10 | — | 98.60% |
 Qwen3.6 35B A3B | akashml/fp8 | 262K | $0.10 | $0.90 | $0.05 | — | 100.00% |
| Qwen3.8 27B | akashml/fp8 | 262K | $0.225 | $1.98 | $0.05 | — | 98.61% |
 gpt-oss-120b | akashml/fp4 | 131K | $0.03 | $0.187 | $0.03 | — | 100.00% |
| gpt-oss-20b | akashml/fp4 | 131K | $0.018 | $0.09 | $0.009 | — | 99.33% |