Relace: Models and Prices
Models and infrastructure for coding agents (apply, embedding and reranking models), and a hosted inference engine.
Relace served 9.54T tokens of other labs' open-weight models over Oct 3 to Oct 9, 2026 through a large public LLM router, #2 among providers.
Open-weight models
| Model | Family | Share of model | Tokens, 7d | Input $/M | Output $/M | Tokens/s | Uptime 3d | Context |
|---|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | DeepSeek | 11.9% | 4.49T | $0.0021 | $0.60 | 115 | 100.0% | 1.05M |
| DeepSeek V4 Flash 0731 | DeepSeek | 31.7% | 1.62T | $0.0063 | $1.28 | 57 | 100.0% | 1.05M |
| GLM 5.3 Flash | GLM | 13.8% | 1.55T | $0.04 | $0.50 | 46 | 100.0% | 1.05M |
| DeepSeek V4 Flash 0423 | DeepSeek | 30.4% | 933B | $0.011 | $1.28 | 91 | 100.0% | 1.05M |
| DeepSeek V4 Pro 0423 | DeepSeek | 33.8% | 203B | $0.30 | $4.20 | 86 | 99.9% | 1.05M |
| Kimi K3 | Kimi | 6.6% | 126B | $2.00 | $14.00 | 64 | 100.0% | 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
- relace.ai
- Latest financing
- $23M Series A led by Andreessen Horowitz (October 2025).
- Status page
- status.relace.ai
Sources: Silicon Legal.
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