Flide launches inference framework for decision models, claims up to 9.5x speedup
What happened
On November 9, Hacker News' front page covered Flide, an inference infrastructure for Jev-style decision models. The report says exact state forking lets a single request respond up to 9.5x faster than vLLM and SGLang, with answers identical to the model answering alone. Flide is served through a /v1/systemone API, supports Qwen3.5, Gemma 4, OneJev and pplx-decider, and is Apache-2.0 licensed.
Written by AI from the coverage · updated 7 hours ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageShow HN: Flide: Make Jev-like decision models up to 9.5x faster
Flide 是面向 Jev 风格决策模型的推理基础设施，借助精确状态分叉，单请求响应速度最高达 vLLM 和 SGLang 的 9.5 倍，且答案与模型单独作答完全一致。它通过 /v1/systemone API 提供服务，已支持 Qwen3.5、Gemma 4 及 OneJev、pplx-decider 等决策模型，采用 Apache-2.0 许可。
Heat over time
Heat now 8·Comparable peak 9（Nov 9 16:00）·Comparable change over 24 hours –
The trend only compares accounts observed without gaps, so its range may be smaller than the current heat. Hover or tap the chart for each hour; the left and right arrow keys step through it.