DekaLLM
Serves 10 models, cheapest of all providers for 1.
Checked Oct 21, 2026
10
models
1
cheapest provider for
98.9%
uptime, last 1 days
95.5%
worst 30-minute reading
Models on DekaLLM
| Model | Input / 1M | Output / 1M | Context | Uptime | Price rank |
|---|---|---|---|---|---|
| Qwen3 30B A3B Instruct 2507 | $0.09 | $0.30 | 262K | 99.7% | cheapest of 3 |
| DeepSeek V4 Pro 0813 | $0.88 | $3.80 | 1.05M | 99.9% | 22 others |
| Gemma 4 26B A4B | $0.06 | $0.33 | 262K | 99.8% | 12 others |
| GLM 5.3 | $0.18 | $2.95 | 1.05M | 89.5% | 46 others |
| gpt-oss-120b | $0.03 | $0.18 | 131K | 100% | 23 others |
| gpt-oss-20b | $0.029 | $0.14 | 131K | 100% | 10 others |
| Mistral Nemo | $0.018 | $0.03 | 131K | 100% | 4 others |
| Nemotron 3 Super | $0.08 | $0.45 | 262K | 100% | 2 others |
| Qwen3.6 35B A3B | $0.10 | $1.00 | 262K | 99.9% | 9 others |
| Qwen3.8 27B | $0.04 | $2.50 | 262K | 100% | 19 others |
Uptime as reported by OpenRouter for this provider’s endpoints. All providers · methodology