Every week, I organize 400 to 700 links into roughly 60 categories as part of my ongoing effort to learn about AI. This is my personal notebook, which I enjoy sharing with friends… a hobby and a labor of love, rather than a commercial publication or product.
If you arrived here through a search or shared link, this page collects the links I found for Moonshot for the week ending July 24, 2026.
As part of my learning process, I like to automate the category covers. It gives me a chance to learn Python and APIs.
This week’s cover prompt was written using Claude Opus 4.7, and the image was generated using Gemini 3.1 Flash Image Preview.
Category cover image prompt:
A single glittering chrome rocket arcing across a deep cosmic purple-black sky toward an enormous glowing chrome moon, its exhaust a swirling ribbon of hot magenta, electric orange, acid yellow, lime, turquoise and violet dotted with gold starburst sparkles, with the word MOONSHOT as a huge bold 1970s funk bubble-letter title arcing along the rocket's trajectory in chrome and glitter with stacked multicolor drop shadows, square Afrofuturist psychedelic concert poster composition with vivid stage-light glow and generous negative space.
This Week in Moonshot News
Here’s a quick AI-generated summary by Claude Sonnet 5.5, based on the headlines and excerpts accompanying this week’s links:
- Strong, but not clearly frontier: Epoch's index put Kimi K3 at a new open-weights record, between Opus 4.6 and GPT 5.4. Together AI found K3 Max matching GPT 5.6 Sol Max on DeepSWE at about 55% of the price. Mollick cautioned that people lean on saturated benchmarks and Arena scores, and that a model months behind can look like the future.
- Distillation fight with Washington: The White House said it has information that Moonshot distilled Anthropic's Fable for K3, and the Treasury threatened sanctions. Skeptics asked for evidence, and one pointed out that only 15 days separated the Fable ban removal from K3's release.
- Demand outran capacity: Moonshot paused new Kimi K3 subscriptions after demand pushed close to its GPU limits. Cline said K3 climbed from 0% to 16% of its open-weights token usage in three days.
This summary was generated by Claude Sonnet 5.5 to help you explore the links below. Rest assured, I select, organize, and check the links by hand in Google Sheets, and write the introduction and personal commentary in The Main Newsletters myself each week as a labor of love.
This week's links related to Moonshot
This reads to me as if preparations are being made to ban models like Kimi K3 in the future. I would be very interested in the evidence that leads to the assumption that Fable 5 was distilled for Kimi K3.”
https://x.com/kimmonismus/status/2079950651644051544
Treasury threatens sanctions after White House claims Moonshot distilled Anthropic’s Fable | TechCrunch
https://techcrunch.com/2026/07/22/treasury-threatens-sanctions-after-white-house-claims-moonshot-distilled-anthropics-fable/
We have information that Moonshot AI distilled Anthropic’s Fable for the development of its K3 model. To do this they developed a sophisticated internal platform to conduct large scale distillation against U.S. models, allowing them to quickly switch between multiple methods of”
https://x.com/mkratsios47/status/2079933645888880708
Kimi Work: Next-Gen Desktop AI Agent for Knowledge Workers
https://www.kimi.ai/products/kimi-work
Moonshot’s Kimi K3 scores 156 on the Epoch Capabilities Index (ECI), setting a new open-weights record. This places it between Opus 4.6, and GPT 5.4, which released in February and March 2026 respectively, and just ahead of GPT 5.6 Luna.”
https://x.com/EpochAIResearch/status/2079602012644360382
Kimi K3 is basically Opus 4.8 on ALE-Bench but Inkling and Grok 4.5 are ngmi”
https://x.com/scaling01/status/2079944011914109189
This is one of the benchmarks I am watching, from the UK’s governmental AI security agency. They will test Kimi K3 when the weights are out in a couple of weeks. It will tell us both whether Kimi has caught up with the public frontier & also kick off a TON of cyber discussions.”
https://x.com/emollick/status/2078144326832451998
“Generate a fake, but believable, witty Churchill insult at a party and explain the context. It should be very clever and original” This time, I think GPT 5.6 Sol Pro wins, but Fable is good too, and you could argue for it taking the prize. Kimi & Gemini miss by a mile.”
https://x.com/emollick/status/2080010641905955328
At the moment that everyone is talking about switching models often for cost or sovereignty or optionality or whatever, the most advanced models are growing more and more different from each other. Fable responds very differently than Kimi K3 or Sol, you can’t just plug & play”
https://x.com/emollick/status/2079631873299320915
Fable, Sol Pro, Kimi K3: “write me a short but good poem using the Odyssey as a basis, think Tennyson or Cavafy” I think this is a Fable victory. Kimi’s is literally a blend of Tennyson’s & Cavafy’s poems themes with some odd bits, and Sol is pretty thematically incoherent.”
https://x.com/emollick/status/2079024884315828351
there are only 15 days between fable 5 ban removal and kimi K3 release. i don’t think claiming that K3’s performance comes from fable distillation (even if they did it) makes sense technically”
https://x.com/eliebakouch/status/2079968464626749888
Kimi K3, like Claude, loves drowned cities, ancient apocalypses, and vast dying gods.”
https://x.com/emollick/status/2078345174334181381
Kimi K3 is a very good model, but people are overindexing on an Arena score again (remember Llama 4?) ELO scores as judged by Arena users are limited, and front-end is like text chat, relatively easy to train/system prompt to a state that people prefer when it is subjective.”
https://x.com/emollick/status/2077969350573572490
We analyzed Kimi K3 Max vs. GPT 5.6 Sol Max for software engineering tasks using DeepSWE. Kimi K3 Max matches GPT 5.6 Sol Max at ~55% of the price. Interestingly – used together, the two models deliver a ~16% performance lift. More insights in the thread! 👇”
https://x.com/togethercompute/status/2080054904328986999
A lot of swift conclusions are being drawn about Kimi K3 based on fairly saturated benchmarks and ELOs, rather than actually testing it on very hard problems. The AI frontier has already moved so far that a good model that is a still months behind looks like the future to many.”
https://x.com/emollick/status/2078129219691798953
Though I would suspect that models like Kimi K3 & GLM-5.2 would also qualify, this is the first time that an open model has reported gold-medal level status at the IMO, which was a rather big threshold when it was crossed last year by (then unreleased) closed models.”
https://x.com/emollick/status/2079944833599156569
Moonshot AI Plans Hong Kong IPO After Kimi K3 Model Debut
https://finance.yahoo.com/markets/stocks/articles/moonshot-ai-plans-hong-kong-123000193.html
remarkable Moonshot is the first model that threatens *spending* on Western closed source, not just token volume. That’s because it’s expensive, verbose and yet great.”
https://x.com/teortaxesTex/status/2079839053483033051
Kimi K3 hit a GPU limit That’s why we saw its sellout that says a lot about where the real bottlenecks are Model capability (yesterday) → Compute/GPUs (today) → Permission (next?)”
https://x.com/TheTuringPost/status/2079727735530815953
I have no inside information, but every sign so far is that there is growing tension about open weights models between the US & China: the US announcing that they reserve the right to act against distilled models, then saying Kimi distills. Contradictory reports from China, etc.”
https://x.com/emollick/status/2080002340497568118
Kimi K3 has become the #3 most used open weights model in ClinePass, going from 0% → 16% token usage in 3 days. This is the fastest climb we’ve seen in open weights usage in Cline’s history.”
https://x.com/cline/status/2080038876929024463
Kimi K3 has received far more love than we expected, and our GPUs are feeling it. Over the past 48 hours, demand has pushed close to the limits of our current capacity. To protect the experience of existing subscribers, we’re temporarily pausing new subscriptions and”
https://x.com/Kimi_Moonshot/status/2078855608565207130?s=20
Kimi K3 needs at least 64 accelerators to deploy. Most people will never run it themselves. But its weights, outputs, and ideas can still shape future models ‒ as Kimi K2.5’s synthetic data helped train @thinkymachines Inkling (one of the biggest US open-weight models) So”
https://x.com/TheTuringPost/status/2079024757031174503
The secret Trump administration battle to fight Chinese AI
https://www.axios.com/2026/07/20/ai-us-china-open-source-kimi
When I asked Kimi K3 “I want you to suggest two poems that you think apply to the current state of GenAI models like you. Don’t just pick popular poems. Think hard” the CoT was 32 pages long (& interesting):
https://t.co/LBPidB5g8a Also typical of K3, lots of looping & dead ends”
https://x.com/emollick/status/2078719596849189323
On Kimi K3: Its Capabilities And Related Discontents | Don’t Worry About the Vase
https://thezvi.wordpress.com/2026/07/20/on-kimi-k3-its-capabilities-and-related-discontents/
Interestingly, when I made a request in Chinese for Kimi K3 to pick two non-cliched poems that apply to LLMs, 95.5% of the characters (88% of the words) in the chain-of-thought were in English, even when it was explicitly considering Chinese poems for a Chinese reader.”
https://x.com/emollick/status/2078621842508587318
Kimi K3’s Design Secret may be in its Thinking Traces
https://notes.designarena.ai/kimi-k3s-design-secret-may-be-in-its-thinking-traces/
There are no frontier open weights models that are not made in China, and there is no incentive in the US, nor appetite in the EU, to build one – it is a lot of cost, little value capture (There are solid mid-level models, of course, but nothing close to a Kimi K3 or a GLM-5.2)”
https://x.com/emollick/status/2079285757991068119
Leave a Reply