Newsroom
Automated, source-cited observations from OpenRouter usage rankings, Hacker News, Hugging Face, and GitHub — trend data, not editorial recommendations. No LLM writes any of this: every entry is a template filled in from the linked source, and every entry cites where the numbers came from.
affaan-m/ECC adds 10,810 stars since our last report
246,692 GitHub stars, up 707 since the last run, active development in the last 30 days. Up 10,810 stars since we covered it on 2026-07-30. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
I trained a small transformer in 1.5hrs and it beats many LLMs is trending on Hacker News
Trending on Hacker News with 622 points and 158 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
minimax/minimax-m3-20260531 adds 8,106,672 monthly requests since our last report
Ranked #20 overall on OpenRouter by request volume. Up 8,106,672 monthly requests since we covered it on 2026-07-28. High OpenRouter rank reflects real, measured API usage, and it is currently climbing.
z-ai/glm-5.2-20260616 climbs OpenRouter rankings
Ranked #20 overall on OpenRouter by request volume, -12% vs. the prior snapshot. Tool-call share ranks #9 in its category. High OpenRouter rank reflects real, measured API usage, and it is currently climbing. Strong tool-call share also suggests agent/tool-use relevance.
TauricResearch/TradingAgents is gaining traction on GitHub
101,799 GitHub stars, up 3,416 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
DietrichGebert/ponytail adds 13,232 stars since our last report
116,769 GitHub stars, up 798 since the last run, active development in the last 30 days. Up 13,232 stars since we covered it on 2026-08-16. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
openai/gpt-oss-20b adds 28,368,360 monthly requests since our last report
Ranked #19 overall on OpenRouter by request volume, +85% vs. the prior snapshot. Up 28,368,360 monthly requests since we covered it on 2026-07-30. High OpenRouter rank reflects real, measured API usage, and it is currently climbing.
I accidentally turned LLM memory into program analysis is trending on Hacker News
Trending on Hacker News with 180 points and 45 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Graphify-Labs/graphify adds 18,067 stars since our last report
111,503 GitHub stars, up 870 since the last run, active development in the last 30 days. Up 18,067 stars since we covered it on 2026-07-25. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
firecrawl/firecrawl adds 12,678 stars since our last report
173,153 GitHub stars, up 660 since the last run, active development in the last 30 days. Up 12,678 stars since we covered it on 2026-08-04. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
browser-use/browser-use is gaining traction on GitHub
111,419 GitHub stars, up 864 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
rohitg00/ai-engineering-from-scratch adds 5,688 stars since our last report
49,264 GitHub stars, up 801 since the last run, active development in the last 30 days. Up 5,688 stars since we covered it on 2026-08-03. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
I were 17, I'd learn how to build LLMs from scratch is trending on Hacker News
Trending on Hacker News with 536 points and 620 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
NousResearch/hermes-agent adds 10,655 stars since our last report
235,255 GitHub stars, up 773 since the last run, active development in the last 30 days. Up 10,655 stars since we covered it on 2026-08-03. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
Why your local LLM feels dumber than it is is trending on Hacker News
Trending on Hacker News with 284 points and 99 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Vomit: Clean up Claude 5's token output with a separate LLM is trending on Hacker News
Trending on Hacker News with 236 points and 241 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
superwhisper/s1-mini is trending on Hacker News
Trending on Hacker News with 107 points and 0 comments. Corroborated by huggingface. A live Hacker News discussion is corroborated by another source right now — developer interest, not a one-off spike.
bojieli/ai-agent-book adds 10,382 stars since our last report
39,887 GitHub stars, up 610 since the last run, active development in the last 30 days. Up 10,382 stars since we covered it on 2026-08-02. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
unslothai/unsloth is gaining traction on GitHub
72,864 GitHub stars, up 689 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
What happens when an LLM never sees material beyond fifth grade? is trending on Hacker News
Trending on Hacker News with 239 points and 206 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
harry0703/MoneyPrinterTurbo is gaining traction on GitHub
104,960 GitHub stars, up 931 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
FareedKhan-dev/train-llm-from-scratch is gaining traction on GitHub
9,140 GitHub stars, up 609 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
Qwen/Qwen3.8-27B is trending on Hugging Face
Trending on Hugging Face with 9914 likes and 91,917 downloads. A real, measured jump in Hugging Face likes/downloads — rising developer interest.
DietrichGebert/ponytail adds 12,255 stars since our last report
103,537 GitHub stars, up 660 since the last run, active development in the last 30 days. Up 12,255 stars since we covered it on 2026-08-02. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
What sort of maths are LLMs good at? is trending on Hacker News
Trending on Hacker News with 254 points and 141 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Apple Silicon and macOS VMs: Faster LLM Inference with llama.cpp is trending on Hacker News
Trending on Hacker News with 290 points and 43 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Stealing Reasoning Traces from Proprietary LLM APIs is trending on Hacker News
Trending on Hacker News with 603 points and 272 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Show HN: Needle2: 14MB agentic LLM for phones, wearables, smart home and robots is trending on Hacker News
Trending on Hacker News with 300 points and 109 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Mark Zuckerberg attacks 'closed' AI rivals as Meta returns to open models is trending on Hacker News
Trending on Hacker News with 483 points and 439 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Humanising LLM Outputs Is Dumb is trending on Hacker News
Trending on Hacker News with 200 points and 131 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
HKUDS/DeepTutor is gaining traction on GitHub
34,197 GitHub stars, up 757 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
How I use LLMs to learn complex topics is trending on Hacker News
Trending on Hacker News with 608 points and 364 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
ZhuLinsen/daily_stock_analysis is gaining traction on GitHub
61,538 GitHub stars, up 696 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
U.S. Department of Energy Launches the Genesis Open Models Initiative is trending on Hacker News
Trending on Hacker News with 205 points and 67 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Born Against, or why hobby programming communities are against LLM usage is trending on Hacker News
Trending on Hacker News with 256 points and 239 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Beating GPT-5.6 Sol on retrieval with 100x cheaper open models is trending on Hacker News
Trending on Hacker News with 314 points and 77 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Position: LLMs Can't Jump is trending on Hacker News
Trending on Hacker News with 273 points and 182 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
LLMs reward expertise is trending on Hacker News
Trending on Hacker News with 907 points and 385 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video is trending on Hacker News
Trending on Hacker News with 293 points and 85 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
MiniMaxAI/MiniMax-H3 is trending on Hugging Face
Trending on Hugging Face with 1682 likes and 0 downloads. A real, measured jump in Hugging Face likes/downloads — rising developer interest.
SQLite Critical CVEs or LLM Slop? is trending on Hacker News
Trending on Hacker News with 710 points and 362 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Prevent cognitive debt by manually retyping LLM-generated code is trending on Hacker News
Trending on Hacker News with 476 points and 385 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
firecrawl/firecrawl is gaining traction on GitHub
160,475 GitHub stars, up 730 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
rohitg00/ai-engineering-from-scratch is gaining traction on GitHub
43,576 GitHub stars, up 616 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
NousResearch/hermes-agent is gaining traction on GitHub
224,600 GitHub stars, up 607 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
DietrichGebert/ponytail is gaining traction on GitHub
91,282 GitHub stars, up 684 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
bojieli/ai-agent-book is gaining traction on GitHub
29,505 GitHub stars, up 740 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
moonshotai/Kimi-K3 is trending on Hugging Face
Trending on Hugging Face with 8337 likes and 99,214 downloads. A real, measured jump in Hugging Face likes/downloads — rising developer interest.
LLM Honeypot is trending on Hacker News
Trending on Hacker News with 231 points and 58 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
affaan-m/ECC is gaining traction on GitHub
235,882 GitHub stars, up 724 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
Now is the time to give LLMs access to the ACM digital library is trending on Hacker News
Trending on Hacker News with 159 points and 127 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Using an open model feels surprisingly good is trending on Hacker News
Trending on Hacker News with 273 points and 95 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
A $500 RL fine-tune of a 9B open model beat frontier models on catalog review is trending on Hacker News
Trending on Hacker News with 176 points and 43 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Running a 28.9M parameter LLM on an $8 microcontroller is trending on Hacker News
Trending on Hacker News with 178 points and 39 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
headroomlabs-ai/headroom is gaining traction on GitHub
62,026 GitHub stars, up 693 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
Graphify-Labs/graphify is gaining traction on GitHub
93,436 GitHub stars, up 784 since the last run, active development in the last 30 days. A real, measured jump in GitHub stars — sustained adoption, not just an old, popular repo.
GigaToken: ~1000x faster Language model tokenization is trending on Hacker News
Trending on Hacker News with 498 points and 100 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
Nativ: Run frontier open models locally on your Mac is trending on Hacker News
Trending on Hacker News with 271 points and 87 comments. An unusually large Hacker News discussion — a real engagement spike, not routine chatter.
ollama/ollama is getting attention on GitHub and hackernews
176,483 GitHub stars, active development in the last 30 days. Corroborated by hackernews. GitHub activity and a live Hacker News discussion agree on this right now — an ecosystem/tooling adoption signal.
Baselines being tracked
Current-state facts with no observed change yet — informational, not signals of movement. They'll surface above once a real change is detected.
- z-ai/glm-5.3-20260816 ranks on OpenRouter#9
- upstage/solar-pro4-20260810 ranks on OpenRouter#17 (+45%)
- tencent/hy4-preview-20260827 ranks on OpenRouter#5
- minimax/minimax-m3-20260531:free ranks on OpenRouter#8
- z-ai/glm-5.3-flash-20260826 ranks on OpenRouter#8
- openai/gpt-5.6-sol-20260709 ranks on OpenRouter#9
- stealth/ox-alpha ranks on OpenRouter#7
- google/gemini-3.7-flash-20260813 ranks on OpenRouter#9
- openai/gpt-4.1-mini-2025-04-14 ranks on OpenRouter#8
- inclusionai/ling-3.0-flash-20260723 ranks on OpenRouter#9
- anthropic/claude-opus-5-20260723 ranks on OpenRouter#8
- poolside/laguna-s-2.1-20260720:free ranks on OpenRouter#9
- deepseek/deepseek-v4-flash-20260731 ranks on OpenRouter#4
- tencent/hy3-20260706:free ranks on OpenRouter#1
- stepfun/step-3.7-flash-20260528 ranks on OpenRouter#6
- openai/gpt-5.6-luna-20260709 ranks on OpenRouter#9
- openai/gpt-5.4-nano-20260317 ranks on OpenRouter#20 (+0%)
- nvidia/nemotron-3-ultra-550b-a55b-20260604:free ranks on OpenRouter#9
- openai/gpt-oss-20b ranks on OpenRouter#20 (-13%)
- anthropic/claude-4.6-sonnet-20260217 ranks on OpenRouter#20 (-8%)
- inclusionai/ling-2.6-flash-20260421 ranks on OpenRouter#19 (+123%)
- deepseek/deepseek-v3.2-20251201 ranks on OpenRouter#19 (-8%)
- minimax/minimax-m3-20260531 ranks on OpenRouter#17 (-12%)
- meta-llama/llama-3.1-8b-instruct ranks on OpenRouter#18 (-2%)
- qwen/qwen3-embedding-8b ranks on OpenRouter#15 (-11%)
- google/gemma-4-31b-it-20260402 ranks on OpenRouter#16 (+3%)
- z-ai/glm-5.2-20260616 ranks on OpenRouter#13 (+14%)
- google/gemma-4-26b-a4b-it-20260403 ranks on OpenRouter#14 (-1%)
- openai/text-embedding-3-small ranks on OpenRouter#12 (+3%)
- deepseek/deepseek-v4-pro-20260423 ranks on OpenRouter#11 (-5%)
- openai/gpt-oss-120b ranks on OpenRouter#9 (+4%)
- mistralai/mistral-nemo ranks on OpenRouter#10 (+3%)
- google/gemini-3-flash-preview-20251217 ranks on OpenRouter#7 (+1%)
- google/gemini-3.1-flash-lite-20260507 ranks on OpenRouter#8 (+6%)
- xiaomi/mimo-v2.5-20260422 ranks on OpenRouter#5 (+55%)
- openai/gpt-4o-mini ranks on OpenRouter#6 (-7%)
- tencent/hy3-20260706 ranks on OpenRouter#3 (+88%)
- google/gemini-2.5-flash ranks on OpenRouter#4 (-3%)
- google/gemini-2.5-flash-lite ranks on OpenRouter#2 (+2%)
- deepseek/deepseek-v4-flash-20260423 ranks on OpenRouter#1 (+2%)