Merge pull request #5 from SplittyDev/patch-1
Merge pull request #5 from SplittyDev/patch-1 Fix TileLang link in README
Merge pull request #5 from SplittyDev/patch-1 Fix TileLang link in README
Merge pull request #3 from youkaichao/main Minor typo fix
Merge pull request #1 from zhyncs/main docs: Update README with HuggingFace and SGLang instructions
docs: Update README with HuggingFace and SGLang instructions Added instructions for using HuggingFace and SGLang with Docker.
Merge pull request #969 from youkaichao/rmsnorm act_quant_kernel
fix act_quant_kernel (#968) Signed-off-by: youkaichao
support scale_fmt=ue8m0 (#964) * support scale_fmt=ue8m0 * keep improving Signed-off-by: youkaichao * keep improving Signed-off-by: youkaichao * add clamp min of 1e-4 Signed-off-by: youkaichao * rename…
Merge pull request #903 from yixing1992/main Update README.md for Huawei Ascend NPU support modes
Update README.md for Huawei Ascend NPU support modes
Merge pull request #666 from codinglover222/deepseek-doc-fix fix an args description.
Merge pull request #736 from shihaobai/main Docs: add LightLLM as supported engine
Merge pull request #816 from KPCOFGS/main Update README.md
Merge pull request #720 from xiaokongkong/main modify the explanation of MLA
RT Kai-Fu LeeThe biggest revelation from Deepseek is that Open Source has won. For a 1% difference in performance, it will be difficult for OpenAI to…
RT Kai-Fu LeeDeepSeek is becoming a Windows kernel demanded by businesses, but http://01.AI is aspired to build the Windows system and interface to ignite it. Check…
DeepSeek's API has been experiencing reliability issues. Here are alternative providers you can use.
DeepSeek shocked the field in late 2024 with DeepSeek-V3, then again with R1, a reasoning model trained at a fraction of the budget Western labs spend. Current lines: V3.2-Exp, R1, V3. The most-talked-about Chinese AI lab of the cycle.
Owner: DeepSeek. We have 280 stories indexed for this model, auto-tagged from titles across every tracked source — official announcements, papers, GitHub release notes, and third-party press. The CTA on each card links to the original; the official site is www.deepseek.com.
Related text models: GPT, Claude, Gemini, Gemma, Llama, Mistral, Grok, Qwen.