product · May 21, 2026
Doubleword Lists Async Gemma-4-31B Inference at $4.1K per Billion Tokens Versus Industry Average $5.2K
Share the canonical public link.
Doubleword offers async overnight batch inference for Gemma-4-31B and additional models including DeepSeek-V4-Pro and Qwen3.5 variants as of the May 2026 site update. Per-token pricing for 1B input plus 1B output tokens stands at $4.1K on Doubleword Async compared with Anthropic at $30K, OpenAI at $15.8K and industry average $5.2K. Up to 75 percent savings apply versus real-time inference for high-volume background jobs. Doubleword Async pricing for Gemma-4-31B reaches $4.1K per billion tokens on the May 2026 platform update.
Below validation threshold — auto-passed without scoring