OpenCode bets token efficiency wins, adding Ling 3.0 Flash for free

OpenCode now offers Ling 3.0 Flash for free, days after inclusionAI released the token-efficient model. The move extends a cost-conscious strategy that has already worked once: OpenCode reached $40M ARR by betting developers want choices, not lock-in, according to our earlier report. Now the wager is that token efficiency beats model size.

Ling 3.0 Flash is built around that trade-off. Fewer tokens per task means AI-assisted coding gets cheaper as usage scales, and giving the model away at no cost puts that math directly in developers' hands. The listing, updated within the last two days, does not list a price beyond the free tier. The same efficiency logic is spreading elsewhere: Cursor added an Auto mode that routes requests to cheaper models, per its July update, and Microsoft matches GPT-5.6 in Excel with a cheaper, older-GPU model that uses fewer tokens than rivals, as that report shows.

For OpenCode, the point is adoption. A free, token-efficient option lowers the barrier for developers watching per-request spend, in a market where agents are increasingly billed by the token. It also offers an alternative to the runaway agent bills that enterprises often can't see, which earlier reporting documented.