🔧 Fine-tune large language models efficiently on NVIDIA DGX Spark with LoRA adapters and optimized quantization for high performance.
GitHub repository with 5 stars and 1 forks.
Language: Python
Topics: blackwell, deep-learning, dgx-spark, fine-tuning, llm, lora, mxfp8, nvfp4, nvidia, pytorch
Trending score 0.05, activity score 0.05, stars gained +0, forks gained +0.
2026-06-13: 5 stars and 1 forks.
A high-throughput and memory-efficient inference and serving engine for LLMs
GitHub repository with 82,749 stars and 18,006 forks.
Trending score: 3.88; stars gained: +133; forks gained: +41.
Language: Python
Topics: amd, blackwell, cuda, deepseek, deepseek-v3, gpt
SGLang is a high-performance serving framework for large language models and multimodal models.
GitHub repository with 28,959 stars and 6,510 forks.
Trending score: 2.75; stars gained: +25; forks gained: +32.
Language: Python
Topics: attention, blackwell, cuda, deepseek, diffusion, glm
TokenSpeed is a speed-of-light LLM inference engine.
GitHub repository with 1,422 stars and 154 forks.
Trending score: 2.01; stars gained: +9; forks gained: +4.
Language: Python
Topics: deepseek, gpt-oss, kimi, lightseek, llm, minimax
🔧 Fine-tune large language models efficiently on NVIDIA DGX Spark with LoRA adapters and optimized quantization for high performance.
GitHub repository with 5 stars and 1 forks.
Trending score: 0.05; stars gained: +0; forks gained: +0.
Language: Python
Topics: blackwell, deep-learning, dgx-spark, fine-tuning, llm, lora
Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode.
GitHub repository with 15 stars and 0 forks.
Trending score: 0.02; stars gained: +0; forks gained: +0.
Language: Python
Topics: abliterated, blackwell, fp8, gemma, gemma-4, modelopt
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.
GitHub repository with 25,425 stars and 1,676 forks.
Trending score: 5.73; stars gained: +2,844; forks gained: +202.
Language: Python
Topics: agent, ai, anthropic, compression, context-engineering, context-window
Self-hosted AI workspace.
GitHub repository with 69,801 stars and 8,840 forks.
Trending score: 5.70; stars gained: +951; forks gained: +165.
Language: Python
The agent that grows with you
GitHub repository with 192,442 stars and 33,558 forks.
Trending score: 5.48; stars gained: +990; forks gained: +282.
Language: Python
Topics: ai, ai-agent, ai-agents, anthropic, chatgpt, claude
AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph.
GitHub repository with 66,467 stars and 6,719 forks.
Trending score: 5.25; stars gained: +1,314; forks gained: +109.
Language: Python
Topics: antigravity, claude-code, codex, gemini, graphrag, knowledge-graph
AI generates a real, editable PowerPoint from any document — native shapes & animations, speaker notes voiced as audio narration, and the option to follow your own .pptx template, not slide images · by Hugo He
GitHub repository with 27,112 stars and 2,418 forks.
Trending score: 5.10; stars gained: +903; forks gained: +61.
Language: Python
Topics: ai-agent, aippt, office, powerpoint, powerpoint-generation, ppt
LLM驱动的 A/H/美股智能分析：多数据源行情 + 实时新闻 + LLM决策仪表盘 + 多渠道推送，零成本定时运行，纯白嫖. LLM-powered stock analysis system for A/H/US markets.
GitHub repository with 42,384 stars and 40,172 forks.
Trending score: 4.94; stars gained: +532; forks gained: +367.
Language: Python
Topics: a-stock, ai-agent, aigc, llm, quant, quantitative-finance
A high-throughput and memory-efficient inference and serving engine for LLMs
GitHub repository with 82,749 stars and 18,006 forks.
Trending score: 3.88; stars gained: +133; forks gained: +41.
Language: Python
Topics: amd, blackwell, cuda, deepseek, deepseek-v3, gpt
SGLang is a high-performance serving framework for large language models and multimodal models.
GitHub repository with 28,959 stars and 6,510 forks.
Trending score: 2.75; stars gained: +25; forks gained: +32.
Language: Python
Topics: attention, blackwell, cuda, deepseek, diffusion, glm
TokenSpeed is a speed-of-light LLM inference engine.
GitHub repository with 1,422 stars and 154 forks.
Trending score: 2.01; stars gained: +9; forks gained: +4.
Language: Python
Topics: deepseek, gpt-oss, kimi, lightseek, llm, minimax
High-performance LLM inference engine in C++/CUDA for NVIDIA Blackwell (RTX 5090/5080/5070 Ti, RTX PRO 6000; sm_120). Native NVFP4/GGUF, 270 tok/s decode on Qwen3-Coder-30B MoE. Written entirely by Claude Code.
GitHub repository with 19 stars and 2 forks.
Trending score: 0.17; stars gained: +0; forks gained: +0.
Language: Cuda
Topics: blackwell, cpp, cuda, cuda-graphs, gated-deltanet, gguf
🔧 Fine-tune large language models efficiently on NVIDIA DGX Spark with LoRA adapters and optimized quantization for high performance.
GitHub repository with 5 stars and 1 forks.
Trending score: 0.05; stars gained: +0; forks gained: +0.
Language: Python
Topics: blackwell, deep-learning, dgx-spark, fine-tuning, llm, lora
Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode.
GitHub repository with 15 stars and 0 forks.
Trending score: 0.02; stars gained: +0; forks gained: +0.
Language: Python
Topics: abliterated, blackwell, fp8, gemma, gemma-4, modelopt