Açık kaynak radarı

Havuz

Takip ettiğimiz konularda yükselen açık kaynak projeler — yıldız ivmesi, gerçek geliştirme aktivitesi ve AI değerlendirmesiyle.

repo · sayfa 1 / 1

Sırala

Yıldız/gün (hız)Son gün (★/gün)Son hafta (★/gün)Son ay (★/gün)Toplam yıldız AI önem skoru Bana uyum Commit (28g)Yazar (28g)En yeni

Konu

Tümü AI / LLM18 Veri9 Geliştirici araçları13 Diğer3 SaaS altyapı13 Web11

Geliştirici araçları

Tümü developer-tools 3.397 cli 2.769 self-hosted 2.477 observability 360 benchmark 356 static-analysis 302 git 280 github-actions 253 monitoring 235 documentation 210 testing 172 ci-cd 112 debugging 94

Süz

Yalnız hareketli Yalnız analizli≥ 0 ≥ 100 ≥ 500 ≥ 2.000

Yalnız hareketli, son 28 günde commit almamış repoları gizler (yıldızı olan ama bırakılmış projeler). Commit sayısı henüz ölçülmemiş repolar da gizlenir.

EtiketTüm GitHub etiketleri — repo sahiplerinin koyduğu topic'lergöster

Sırala:

Etiketler yükleniyor…

Aktif speculative-decoding

1
leonxger/deepseek-colab 14
Uncensored DeepSeek-V4.1-Flash IQ2_XS on Google Colab
4,7/gün ort. 1 commit 1 yazar Python 30 Eki 2026 güncellendi 4 gün önce
colab cuda deepseek deepseek-v4 gguf google-colab gpu llama-cpp
14
2
hasso5703/alzeo 245
dgx-spark-qwen38 is an open-source installer that turns an NVIDIA DGX Spark, or any GB10 machine (ASUS, Dell, HP, Lenovo, MSI, GIGABYTE, Ace
3,1/gün ort. +0 son gün 608 commit 1 yazar Python 15 Ağu 2026 güncellendi 2 gün önce
ai-agents asus-ascent-gx10 dgx-spark gb10 image-generation llm-inference local-ai local-llm
245
3
0xBakeer/TandemLLM 70
An environment where open models work as agents: a first-level harness with routing, skills, memory and sandboxed tools, on an engine that d
2/gün ort. +0 son gün — commit Python 28 Eyl 2026 güncellendi dün
ai-agents dgx-spark exl3 inference-engine llm-inference local-ai nvfp4 quantization
70
4
Blackwellboy/model-serving-minefield 165
Community registry of LLM serving-path traps that produce confidently wrong measurements: templates, tool parsers, reasoning fields, quant k
1,7/gün ort. +1 son gün 34 commit 1 yazar Python 27 Tem 2026 güncellendi dün
benchmarking chat-template cuda debugging llama-cpp llm-serving lm-eval local-llm
165
5
raketenkater/ggrun 282
llama.cpp/ik_llama.cpp launcher: loads big MoE models across mismatched multi-GPU rigs by exact VRAM math.
1,2/gün ort. +0 son gün 242 commit 2 yazar Go 11 Mar 2026 güncellendi 2 gün önce
cuda gguf llama-cpp llm metal moe multi-gpu golang
282
6
Lowaltitudeflight/local-llm-tuning-guide 19
A decision library for running local LLMs on multi-GPU machines without NVLink: parallelism, communication, quantization, speculative decodi
1,1/gün ort. +0 son gün — commit Python 16 Eki 2026 güncellendi dün
benchmarking decision-guide gpu-serving inference-optimization kv-cache llm llm-deployment llmops
19
7
swellweb/reame 114
CPU-first LLM inference server on llama.cpp. Runs useful models on free-tier ARM boxes; rewriting the input made it ~6x faster and more accu
1/gün ort. +0 son gün — commit C++ 06 Tem 2026 güncellendi 22 gün önce
arm cpu edge-ai gguf inference kv-cache llama-cpp llm
114
8
FedericoTs/quantprobe 92
Run a 110B on a 2016 PC with 16 GB RAM. Know your tok/s before you download. Placement beats budget: predicts speed + memory fit for any GGU
0,9/gün ort. +0 son gün — commit Python 21 Tem 2026 güncellendi 27 gün önce
benchmark gguf hardware-detection inference llama-cpp llm local-llm moe
92
9
gpillon/ignis 50
OpenAI-compatible inference server for Qwen3.8 on one RTX 5090 / RTX PRO 6000: the 27B at ~170 tok/s (peaks ~370), Flash-Next (125B MoE) at
0,8/gün ort. +1 son gün — commit Rust 01 Eyl 2026 güncellendi dün
blackwell coding-agents cuda inference-engine llm llm-inference local-llm mixture-of-experts
50
10
0xBakeer/Qwen3.8-27B-FP8-on-a-single-DGX-Spark 55
Serving Qwen3.8-27B-FP8 on a single DGX Spark (GB10): 7.88 to 58.5 tok/s single-stream from decode strategy alone, weights untouched. Specul
0,7/gün ort. +0 son gün — commit Python 16 Ağu 2026 güncellendi 30 Eyl 2026
benchmark dflash2 dgx-spark fp8 gb10 llama-cpp llm-inference llm-serving
55
11
alphastorm/omp-ninfer 47
Durable local inference for Oh My Pi: NInfer + Qwen3.8 27B on one RTX 3090/4090/5090, with restart-resumable OpenAI Responses state.
0,7/gün ort. +0 son gün — commit Python 27 Ağu 2026 güncellendi dün
ai-agent coding-agent coding-assistant cuda directstorage local-ai local-llm nvidia
47
12
Sandermage/sndr_core_engine 133
SNDR Core Engine (Genesis) — vLLM runtime patch-overlay for Qwen3.6 + Gemma4 on consumer NVIDIA (Ampere sm_86, 2× A5000/3090). Qwen3.6-35B-A
0,7/gün ort. +0 son gün — commit Python 17 Nis 2026 güncellendi 01 Eki 2026
cuda llm-inference qwen vllm speculative-decoding tool-calling rtx-3090 awq
133
13
0xBakeer/Qwen3.8-27B-4-bit-on-a-single-DGX-Spark 51
Serving 4-bit Qwen3.8-27B on a single DGX Spark (GB10): 75 tok/s single-stream, 246 tok/s aggregate at 8-way concurrency. NVFP4 vs MixedInt4
0,7/gün ort. +0 son gün — commit Python 16 Ağu 2026 güncellendi 30 Eyl 2026
autoround benchmark dgx-spark gb10 int4 llm-inference llm-serving marlin
51
14
aic0d3r/qwen38-strix-halo-harness 37
AI coding agent harness for any hardware, local or cloud (OpenAI-compatible engines). Tuned for AMD Strix Halo: Qwen3.8 125B on the iGPU, NP
0,5/gün ort. +0 son gün — commit TypeScript 23 Ağu 2026 güncellendi 2 gün önce
ai-coding-assistant amd benchmark coding-agent llama-cpp llm lm-studio local-llm
37
15
lifeidle/qwen3.8-flash-next-5090-laptop-24gb-strata 26
Qwen3.8-Flash-Next 177B MoE (IQ3_S 3.44bpw) on RTX 5090 Laptop 24GB: full-model quality @ 97.7 tok/s · dual-lane 256K full-context unlocked
0,5/gün ort. +1 son gün — commit Python 10 Eyl 2026 güncellendi 6 gün önce
256k-context flash-next gguf iq3-xxs local-llm moe mtp quantization
26
16
lifeidle/qwen3.8-27b-5090-laptop-24gb-coder390-ninfer 20
Qwen3.8-27B on RTX 5090 Laptop 24GB — TWIN-TURBO NVFP4 (native llama.cpp) vs Coder390 Q3 LynnStyle (.ninfer + NInfer engine in WSL2/Docker):
0,4/gün ort. +0 son gün — commit Python 10 Eyl 2026 güncellendi 18 gün önce
24gb benchmark blackwell docker gguf kv-cache llama-cpp local-llm
20
17
AEON-7/Aeon-Bench-Pod 26
Run the AEON Bench suite on your own hardware: verified HuggingFace pull → serve → benchmark (text · agentic ×3 harnesses · vision · audio ·
0,2/gün ort. +0 son gün — commit Python 04 Tem 2026 güncellendi 28 gün önce
benchmark dgx-spark evaluation llm nvfp4 speculative-decoding vllm
26
18
nicremo/qwen38-runpod-stack 14
One command spins up Qwen3.8-27B (abliterated, NVFP4) on a rented Blackwell GPU at ~150 tok/s and wires it into your local coding agent via
0,2/gün ort. +0 son gün — commit Shell 23 Ağu 2026 güncellendi 23 Ağu 2026
abliterated blackwell coding-agent gpu inference llm nvfp4 openai-compatible
14
19
thc1006/qwen3.6-speculative-decoding-rtx3090 35
llama.cpp speculative decoding measured on one RTX 3090, Qwen3.6-35B-A3B UD-Q4_K_XL, commit 3737e4137. Published figures are re-derived from
0,2/gün ort. +0 son gün — commit Python 21 Nis 2026 güncellendi 14 gün önce
ampere benchmark cuda ggml inference-benchmark llama-cpp local-llm mixture-of-experts
35
20
zorost/sparkduet 0
Serve DeepSeek and Qwen, run an on-demand model library, and fine-tune with Unsloth QLoRA on 2x NVIDIA DGX Spark. TP=2 vLLM lanes behind one
0/gün ort. +0 son gün — commit Python 25 Ağu 2026 güncellendi 07 Eyl 2026
coding-agents deepseek dgx-spark fine-tuning gb10 homelab llama-swap llm-inference
0