Darkbloom: Models and Prices
Decentralized inference network by Eigen Labs that runs models on idle Apple Silicon Macs.
Darkbloom served 108B tokens of other labs' open-weight models over Oct 3 to Oct 9, 2026 through a large public LLM router, #37 among providers.
Open-weight models
| Model | Family | Share of model | Tokens, 7d | Input $/M | Output $/M | Tokens/s | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| Gemma 4 26B A4B | Gemma | 22.3% | 59B | $0.042 | $0.22 | 35 | 100.0% | 131K |
| gpt-oss-20b | gpt-oss | 16.4% | 33B | $0.018 | $0.09 | 23 | 100.0% | 131K |
| Qwen3.6 35B A3B | Qwen | 18.7% | 7B | $0.05 | $0.70 | 61 | 100.0% | 262K |
| Qwen3.5-35B-A3B | Qwen | 100.0% | 3B | $0.08 | $0.75 | 67 | 100.0% | 262K |
| Qwen3.8 27B | Qwen | 0.4% | 2B | $0.05 | $2.20 | 24 | 99.1% | 262K |
| Qwen3.5-9B | Qwen | 9.6% | 1B | $0.08 | $0.13 | 26 | 99.8% | 262K |
| MiMo-V2.6-Flash | MiMo | <0.1% | 955M | $0.10 | $0.28 | 17 | 98.6% | 1.05M |
| Nemotron 3.5 Lightning | Nemotron | <0.1% | 382M | $0.025 | $0.12 | 59 | 99.6% | 262K |
| Ternary Bonsai 2 27B | Other open-weight | 100.0% | 67M | $0.075 | $0.50 | 26 | 100.0% | 262K |
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
- darkbloom.dev
- Compute
- No data centers: capacity comes from Mac owners who enroll their machines.
Sources: Darkbloom docs.
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