Fortytwo launches Engine, an inference API it says gets better answers from open models using swarm inference
Introducing Fortytwo Engine An inference API that runs open-weight models through Swarm Inference • 7% higher scores on average vs. the base model • Nearly 25% fewer tokens per task • Same per-token price and speed • Private by default First model: Qwen3.8-27B Try it:
Context
If running many copies of an open model together reliably beats the single model at the same price, it could make cheap open models more competitive.
This is a promotional launch post from Fortytwo, and all benchmark gains are the company's own measurements. Only one model, Qwen3.8 27B, is offered so far.
Deep dive
On October 15, Fortytwo posted: Introducing Fortytwo Engine, an inference API that runs open weight models through what it calls Swarm Inference. It claimed 7% higher scores on average than the base model, nearly 25% fewer tokens per task, the same per token price and speed, and privacy by default, with Qwen3.8 27B as the first model.
The product page lists Fortytwo's own benchmark results for Qwen3.8 27B through Engine compared with the base model, including DeepSWE rising from 42.2 to 47.8, GPQA Diamond from 90.9 to 95.1 and LiveCodeBench v6 from 80.2 to 84.1. It says average tokens per run across 450 DeepSWE runs fell from 16.3 million to 12.3 million.
The page lists pricing of $0.425 per million input tokens and $2.55 per million output tokens, a context window of 262,144 tokens, zero data retention and an endpoint compatible with the standard chat completions format.
The page does not explain in detail how swarm inference produces better answers at the same cost, and no independent tests were found.
What to expect next
Watch for independent benchmarks that confirm whether Engine beats the plain Qwen model under the same conditions and price.
Also watch which open models Fortytwo adds next, how its speed holds up under real load, and whether developers adopt it over standard hosts for open weight models. Pricing changes after launch will also matter for developers comparing it with other hosts.