Pricing, side by side
Run BiOS vs Fireworks, Together AI and Nebius
What the same frontier models cost per million tokens on Run BiOS, Fireworks, Together AI and Nebius — blended at 70% input / 30% output, the mix most chat and coding workloads sit near.
| Model | Run BiOS / 1M | Fireworks | Together AI | Nebius |
|---|---|---|---|---|
| DeepSeek V4 Flashdeepseek-v4-flash | $0.15 | $0.18−20% | not listed | not listed |
| Kimi K2.7 Codekimi-k2.7-code | $1.58 | $1.86−15% | $1.86−15% | $1.86−15% |
| GLM 5.2glm-5.2 | $2.00 | $2.30−13% | $2.30−13% | $2.30−13% |
| DeepSeek V4 Prodeepseek-v4-pro | $2.00 | $2.26−12% | $2.26−12% | not listed |
| Qwen3.5 397Bqwen3.5-397b-a17b | $1.37 | not listed | $1.50−9% | $1.50−9% |
| MiniMax M3minimax-m3 | $0.57 | $0.57 | $0.57 | $0.57 |
Blended USD per 1M tokens at 70% input / 30% output, Run BiOS rates as of 17 August 2026. Fireworks: Standard tier, from its pricing documentation 26 July 2026. Together AI: serverless list prices, 26 July 2026. Nebius: Token Factory list, 26 July 2026. Models a provider does not price individually show “not listed” — we do not guess a rival’s rate. Run BiOS wins on 5 of the 6 models above; where we do not, the table says so. The exact rate for the model you are about to call is shown in the dashboard before you send a request — treat that as authoritative.