AkashML: Models and Prices
Inference API on the Akash decentralized compute network.
AkashML served 89B tokens of other labs' open-weight models over Oct 3 to Oct 9, 2026 through a large public LLM router, #40 among providers.
Open-weight models
| Model | Family | Share of model | Tokens, 7d | Input $/M | Output $/M | Tokens/s | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| gpt-oss-120b | gpt-oss | 11.9% | 50B | $0.03 | $0.187 | 91 | 99.9% | 131K |
| Qwen3.6 35B A3B | Qwen | 36.6% | 14B | $0.10 | $0.90 | 76 | 99.4% | 262K |
| gpt-oss-20b | gpt-oss | 3.8% | 8B | $0.02 | $0.10 | 32 | 99.9% | 131K |
| GLM 5.3 | GLM | 0.2% | 7B | $0.19 | $4.40 | 79 | 100.0% | 1.05M |
| Llama 3.3 70B Instruct | Llama | 40.5% | 6B | $0.20 | $0.52 | 34 | 99.6% | 131K |
| Qwen3.8 27B | Qwen | 0.5% | 3B | $0.225 | $1.98 | 52 | 99.7% | 262K |
| Kimi K3 | Kimi | <0.1% | 738M | $1.20 | $14.00 | 63 | 98.9% | 1.05M |
List prices per million tokens as of October 9, 2026. Share: of all tokens served on that model; a dash where the volume is not reported.
Company
- Website
- akashml.com
Sources: AkashML.
Finance your GPUs
Serving tokens on your own GPUs? Tell us the cluster, the models you serve and the contracts behind them, and we will share it with partner funders that fit.
Prefer email? hello@amcompute.com