Fireworks AI(@FireworksAI_HQ)
Qwen 3.5 from @Alibaba_Qwen is now available on @FireworksAI_HQ Training Platform across the Managed...
7.2Score
TL;DR · AI 摘要
Fireworks AI 宣布 Qwen 3.5 正式接入其训练平台，支持 SFT/DPO/RLHF 等全栈微调方式、256K 上下文及 LoRA/全参微调。
核心要点
- Qwen 3.5 已上线 Fireworks AI 训练平台，覆盖托管与 API 两种工作流
- 支持 SFT、DPO、强化学习（RL）微调，提供智能默认配置与自定义损失函数能力
- 具备 256K 超长上下文窗口，并同时兼容 LoRA 与全参数微调
结构提纲
按章节快速跳转。
思维导图
用一张图看清主题之间的关系。
查看大纲文本（无障碍 / 无 JS 友好）
- Qwen 3.5 × Fireworks AI 训练平台
- 训练范式
- SFT（监督微调）
- DPO（直接偏好优化）
- RL（强化学习）
- 关键技术特性
- 256K 上下文窗口
- LoRA / 全参微调双支持
- 自定义损失函数接口
- 部署方式
- Managed 工作流
- Training API 工作流
金句 / Highlights
值得收藏与分享的关键句。
Qwen 3.5 from @Alibaba_Qwen is now available on @FireworksAI_HQ Training Platform across the Managed and Training API workflows.
Try SFT, DPO, RL with smart defaults or your own custom loss function with a 256K context window.
We support Lora as well as full param fine tuning for your most advanced tasks!
#Qwen#Fireworks AI#大模型微调#LLM#Alibaba
打开原文Try SFT, DPO, RL with smart defaults or your own custom loss function with a 256K context window. We support Lora as well as full param fine tuning" / X
Fireworks AI on X: "Qwen 3.5 from @Alibaba_Qwen is now available on @FireworksAI_HQ Training Platform across the Managed and Training API workflows. Try SFT, DPO, RL with smart defaults or your own custom loss function with a 256K context window. We support Lora as well as full param fine tuning" / X
Don’t miss what’s happening
Qwen 3.5 from
is now available on
Training Platform across the Managed and Training API workflows. Try SFT, DPO, RL with smart defaults or your own custom loss function with a 256K context window. We support Lora as well as full param fine tuning for your most advanced tasks! What would you like to see next?
·
2
1
36
8