AIトレンド・ダッシュボード

最終更新: 2026-07-05（UTC基準）

この一覧は自動収集した数値と、AIが生成した日本語1行解説（データノート）です。解説は未レビューの参考情報であり、数値は各一次ソースに基づきます。（本日の自動解説は未生成のため、数値のみを表示しています）

GitHub スター（追跡リポジトリ）

#	リポジトリ	Stars	Forks	License
1	n8n-io/n8n	195,316	59,084	NOASSERTION
2	Significant-Gravitas/AutoGPT	185,389	46,119	NOASSERTION
3	ollama/ollama	175,549	16,854	MIT
4	AUTOMATIC1111/stable-diffusion-webui	163,997	30,378	AGPL-3.0
5	huggingface/transformers	162,284	33,778	Apache-2.0
6	langgenius/dify	147,801	23,274	NOASSERTION
7	open-webui/open-webui	144,339	20,859	NOASSERTION
8	langchain-ai/langchain	141,018	23,436	MIT
9	Comfy-Org/ComfyUI	119,551	14,035	GPL-3.0
10	ggml-org/llama.cpp	119,380	20,258	MIT
11	openai/whisper	104,266	12,708	MIT
12	modelcontextprotocol/servers	88,093	11,160	NOASSERTION
13	vllm-project/vllm	85,441	18,989	Apache-2.0
14	lobehub/lobehub	79,494	15,562	NOASSERTION
15	unslothai/unsloth	67,829	6,104	Apache-2.0
16	microsoft/autogen	59,511	8,962	CC-BY-4.0
17	crewAIInc/crewAI	54,968	7,716	MIT
18	FlowiseAI/Flowise	54,302	24,649	NOASSERTION
19	BerriAI/litellm	52,684	9,497	NOASSERTION
20	ggml-org/whisper.cpp	51,320	5,722	MIT

Hugging Face（DL数上位）

#	モデル/データ	種別	DL数	Likes
1	sentence-transformers/all-MiniLM-L6-v2	model	244,099,047	5,030
2	cross-encoder/ms-marco-MiniLM-L6-v2	model	79,982,687	272
3	google-bert/bert-base-uncased	model	62,774,318	2,696
4	BAAI/bge-small-en-v1.5	model	61,557,234	503
5	sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2	model	47,685,110	1,298
6	google/electra-base-discriminator	model	39,011,105	132
7	sentence-transformers/all-mpnet-base-v2	model	33,023,896	1,318
8	BAAI/bge-m3	model	32,338,205	3,191
9	Qwen/Qwen3-0.6B	model	27,835,708	1,384
10	timm/mobilenetv3_small_100.lamb_in1k	model	26,534,440	83
11	openai/clip-vit-base-patch32	model	22,262,188	969
12	FacebookAI/xlm-roberta-base	model	20,506,108	860
13	google-t5/t5-small	model	19,836,097	564
14	Qwen/Qwen3-8B	model	16,401,211	1,182
15	nomic-ai/nomic-embed-text-v1.5	model	16,250,819	859

Hacker News（AIトピック話題度）

#	タイトル	Points	Comments
1	Jamesob's guide to running SOTA LLMs locally [HN]	405	181
2	Show HN: Bramble – Local-first password manager [HN]	142	43
3	Show HN: KiCad in the Browser [HN]	95	32
4	The Plight of the Martian Farmer [HN]	35	9
5	Does Code Cleanliness Affect Coding Agents? [HN]	23	3
6	Connections in Math: the two kinds of random [HN]	20	9
7	Tell HN: don't trust Bigco AI agents with AI research IP [HN]	16	6
8	Context graphs: how AI agents can store and use past decisions [HN]	9	0
9	Building Agents That Don't Break Themselves [HN]	8	0
10	Show HN: Sidenote – comment on your rendered blog, an LLM writes the Git diff [HN]	7	0
11	Show HN: I hated how much my 12-year-old played Roblox, so we built our own FPS [HN]	7	0
12	Show HN: Pull Claude Code transcripts into your Codex session, and vice versa [HN]	7	1
13	Ask HN: Best Local LLM Setup for a 128GB M4 Max Mac Studio? [HN]	7	0
14	Ask HN: Is anyone running local LLMs in their organization? [HN]	7	0
15	OpenAI is fast-tracking its own "AI Agent Phone" for 2027 to challenge iPhone [HN]	5	3

npm / PyPI（直近週DL数）

#	パッケージ	レジストリ	週間DL
1	langchain	pypi	74,245,269
2	transformers	pypi	41,814,673
3	anthropic	pypi	37,090,294
4	openai	npm	25,115,784
5	ai	npm	15,124,270
6	@langchain/core	npm	4,585,330
7	langchain	npm	2,373,902
8	@huggingface/inference	npm	380,277
9	llamaindex	npm	119,719

arXiv 新着（cs.AI / cs.CL / cs.LG）

タイトル	カテゴリ	公開日
Distributed Attacks in Persistent-State AI Control	cs.AI	2026-07-02
LACUNA: A Testbed for Evaluating Localization Precision for LLM Unlearning	cs.CL, cs.AI, cs.LG	2026-07-02
Program-as-Weights: A Programming Paradigm for Fuzzy Functions	cs.LG, cs.AI, cs.CL	2026-07-02
Online Safety Monitoring for LLMs	cs.AI, cs.CL, cs.LG, stat.AP, stat.ML	2026-07-02
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning	cs.AI	2026-07-02
What LLM Agents Say When No One Is Watching: Social Structure and Latent Objective Emergence in Multi-Agent Debates	cs.AI, cs.CL, cs.LG, cs.MA	2026-07-02
Reasoning LLM Improves Speaker Recognition in Long-form TV Dramas	cs.CL, cs.AI, cs.CV	2026-07-02
DemoPSD: Disagreement-Modulated Policy Self-Distillation	cs.LG, cs.AI	2026-07-02
Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials	cs.LG, cs.AI, physics.chem-ph, physics.comp-ph	2026-07-02
G-RRM: Guiding Symbolic Solvers with Recurrent Reasoning Models	cs.AI	2026-07-02
Combating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning	cs.CV, cs.AI	2026-07-02
TestEvo-Bench: An Executable and Live Benchmark for Test and Code Co-Evolution	cs.SE, cs.AI, cs.CL	2026-07-02
Human Capital, Not Model Benchmarks, Predicts Hybrid Intelligence in Forecasting	cs.CY, cs.AI	2026-07-02
Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs	cs.RO, cs.AI	2026-07-02
OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers	cs.CV, cs.AI, cs.LG	2026-07-02
Neuron-Aware Data Selection for Annotation-Free LLM Self-Distillation	cs.LG, cs.AI	2026-07-02
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments	cs.AI, cs.CL	2026-07-02
Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study	cs.SE, cs.AI	2026-07-02
Automated grading of Linux/bash examinations using large language models: a four-level cognitive taxonomy approach	cs.AI, cs.CL, cs.CY	2026-07-02
WorldSample: Closed-loop Real-robot RL with World Modelling	cs.RO, cs.AI	2026-07-02