InnerVault

Library

85,340 resources from 11 public sources. 54 checked by hand; the rest are unreviewed.

Results

  1. vllm

    AI repositories

    A high-throughput and memory-efficient inference and serving engine for LLMs

    Indexed GitHub, vllm-project 97.0k ★Apache-2.0Updated today

    git clone https://github.com/vllm-project/vllm.git
  2. sglang

    AI repositories

    SGLang is a high-performance serving framework for large language models and multimodal models.

    Indexed GitHub, sgl-project 38.9k ★Apache-2.0Updated today

    git clone https://github.com/sgl-project/sglang.git
  3. openlake

    AI repositories

    OpenLake is a high performance storage engine for efficient LLM inference and GPU Training

    Stale GitHub, openlake-project 2,851 ★Apache-2.0Updated 7 days ago

    git clone https://github.com/openlake-project/openlake.git
  4. tokenspeed

    AI repositories

    TokenSpeed is a speed-of-light LLM inference engine.

    Indexed GitHub, lightseekorg 2,932 ★MITUpdated today

    git clone https://github.com/lightseekorg/tokenspeed.git
  5. image text to text model by autotrust · vllm

    Stale Hugging Face Hub, autotrust 561 ♥mitUpdated 36 days ago

    autotrust/GLM5.3-Flash-E224-DGX-Spark
  6. image text to text model by autotrust · vllm

    Stale Hugging Face Hub, autotrust 180 ♥mitUpdated 36 days ago

    autotrust/GLM5.3-Flash-E192-DGX-Spark
  7. text generation model by BlackwellBoy

    Stale Hugging Face Hub, BlackwellBoy 25 ♥mitUpdated 26 days ago 1 note

    BlackwellBoy/DeepSeek-V4.1-Flash-1017-BLACKFOLD-EXL3-2.9bpw-TP2
  8. image text to text model by autotrust · vllm

    Stale Hugging Face Hub, autotrust 118 ♥mitUpdated 36 days ago

    autotrust/GLM5.3-Flash-E256-DGX-Spark
  9. ds4-on-spark

    Datasets & benchmarks

    Entrpi/ds4, a Blackwell CUDA perf fork of antirez/ds4 on NVIDIA DGX Spark: one-command install, ~3x upstream prefill, ~1.5x decode, DSpark, and full continuous batch support

    Stale GitHub, Entrpi 409 ★MITUpdated 2 months ago

    git clone https://github.com/Entrpi/ds4-on-spark.git
  10. agent-lsp

    MCP servers

    Language servers as 66 code-intelligence tools for AI agents, CI-verified on 32 languages.

    Stale Official MCP Registry, io.github.blackwell-systems 170 ★Updated 12 days ago 1 note

    @blackwell-systems/agent-lsp@0.24.2
  11. imp

    Agents & frameworks

    Run large AI chat models privately on your own NVIDIA RTX 50 or RTX PRO graphics card. On one RTX 5090, Qwen3.8-27B writes about 70 words per second, 1.4x faster than vLLM. Even a model almost four times larger than the card's memory runs. Works with apps made for ChatGPT or Claude. Starts with one docker command.

    Stale GitHub, kekzl 124 ★MITUpdated 12 days ago

    git clone https://github.com/kekzl/imp.git
  12. knowing

    MCP servers

    Content-addressed code graph with 22 MCP tools for AI agents.

    Stale Official MCP Registry, io.github.blackwell-systems Updated 54 days ago 1 note

    @blackwell-systems/knowing@0.16.0
  13. gcf-proxy

    MCP servers

    Drop-in MCP proxy. 71% fewer tokens. Session dedup compounds to 92%. Zero code changes.

    Stale Official MCP Registry, io.github.blackwell-systems Updated 5 months ago 1 note

    @blackwell-systems/gcf-proxy@0.10.2
  14. gcf

    MCP servers

    The AI-native wire format for structured data. 100% comprehension on every frontier model. 50-92% fewer tokens than JSON. 43B+ lossless round-trips across 17 formats. Spec v3.5.1 Stable.

    Indexed GitHub, blackwell-systems 61 ★Apache-2.0Updated today

    git clone https://github.com/blackwell-systems/gcf.git
  15. mcp-assert

    MCP servers

    The testing standard for MCP servers. Lint (14 rules), test, and fuzz over real stdio/SSE/HTTP transport. 18 assertion types in YAML. Found 4,794 schema issues across 55 servers. Any language, no mocks, single binary.

    Indexed GitHub, blackwell-systems 26 ★MITUpdated today

    git clone https://github.com/blackwell-systems/mcp-assert.git

Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.