Highlights
- Pro
Pinned Loading
-
jev-native-agent-with-extended-options
jev-native-agent-with-extended-options PublicResearch design for a Jev-native agent system:enable more options than jev provided with virtulization and paging, tool integration, external helper logits Top-k proposals with Jev-controlled fallb…
-
elastic-kv-vmm
elastic-kv-vmm PublicDemand-backed CUDA VMM KV cache prototype for SGLang, with real A100 serving measurements and allocator ablations.
Python
-
selection-prefix-cache
selection-prefix-cache PublicResearch prototype for compressed sparse-selection replay with bounded CUDA prefetch and reusable buffers.
Python
-
strict-topk-overlap
strict-topk-overlap PublicStrict Top-K speculative attention: guarded hybrid repair, Triton kernels and reproducible A100 experiments.
Python
-
dsh-wsl-workspace-maintainers/dsh-wsl-workspace
dsh-wsl-workspace-maintainers/dsh-wsl-workspace PublicWSL workspace support for DeepSeek Harness——无缝的 WSL 工作区使用体验,无需在 WSL 之中再安装一个dsh,安装该插件后在 GUI 里直接添加 WSL 工作区即可。WSL workspace support for DeepSeek Harness — Enjoy a seamless WSL workspace experience wit…
-
request-layer-aware-hisparse
request-layer-aware-hisparse PublicResearch proposal for request- and layer-aware HiSparse hot buffers
If the problem persists, check the GitHub status page or contact support.



