lreeves

This is completely untrue and was based on some weird TikTok rumors.

But won't setting it to accept 100% of the proposed tokens will skip the verification?

Doesn't accepting 100% of the MTP draft tokens mean you should just be using the smaller model? Usually the acceptance rate in Qwen36 at least is around 60-70% and the "wrong" tokens are still filled in entirely by the base model, but when you just accept 100% of the draft tokens it seems kind of self defeating unless I'm wrong.

Also I feel like everyone leaves off prompt processing/prefill speeds in these articles. If you are using a very small prompt and asking for mostly generated tokens, sure but I'd love to know the time-to-response of asking for an analysis of an image or a few hundred lines of code.

You showed this setup to a business insurance underwriter and they gave you a policy? Can I ask how much the premium is? Or is this just theft insurance?

I'm running the unquantized base model on 2xA6000s (Ampere gen, 48GB each). Runs at about 25 tokens/second.

I have been kicking the tires for about 40 minutes since it downloaded and it seems excellent at general tasks, image comprehension and coding/tool-calling (using VLLM to serve it). I think it squeaks past Gemma4 but it's hard to tell yet.

I bought Battlefield 2 and it's DLC and one of the earlier Dirt games on EA Origin and it was an absolute nightmare. My games and the DLC would constantly not be authed in my account and I still have like dozens of support threads in my old mailbox trying to get things working.

At the same time Steam had polished a lot of the rough edges like this for their catalog and other publishers so there's really no excuse. I've never had to open support tickets with any other storefront because the DLC map pack for a game would stop loading while the base game kept working.

In my experience Qwen3.5/Qwen3-Coder-Next perform best in their own harness, Qwen-Code. You can also crib the system prompt and tool definitions from there though. Though caveat, despite the Qwen models being the state of the art for local models they are like a year behind anything you can pay for commercially so asking for it to build a new app from scratch might be a bit much.

Could not agree more, it's wild my AppleTVs now ask for which profile is using it but the iPad still hasn't gotten this.

I mean no shit though? People calmly said this in Trump's first term where he (unsuccessfully) first tried to go tariff crazy. What does it add though? Nobody is freaking out saying "all tariffs are bad", they're saying "blanket tariffs for no/the stupidest reasons possible are bad".

I run the larger version of it on a Threadripper with 512GB RAM and a 32GB GPU for the non-expert layers and context, using llama.cpp. Performs great, however god forbid you try to get that much memory these days.

I sometimes still code with a local LLM but can't imagine doing it on a laptop. I have a server that has GPUs and runs llama.cpp behind llama-swap (letting me switch between models quickly). The best local coding setup I've been able to do so far is using Aider with gpt-oss-120b.

I guess you could get a Ryzen AI Max+ with 128GB RAM to try and do that locally but non-nVidia hardware is incredibly slow for coding usage since the prompts become very large and take exponentially longer but gpt-oss is a sparse model so maybe it won't be that bad.

Also just to point it out, if you use OpenRouter with things like Aider or roocode or whatever you can also flag your account to only use providers with a zero-data retention policy if you are truly concerned about anyone training on your source code. GPT5 and Claude are infinitely better, faster and cheaper than anything I can do locally and I have a monster setup.

It's not practical when Trump sees a TV ad that enrages him and then cancels all negotiations, how are Canadian leaders supposed to proceed? There's no good faith whatsoever from him.

Exactly right. There are trade deals forming between countries that in unprecedented ways to avoid dealing with the constantly changing tariffs while one country says they'll take their ball and play alone.

In a sense you can think of it that way, as a Canadian we counter-tariff the US and that can be considered punishing us; however the US is only one country and it encouraged more free trade with every other one of our trading partners so in a game theory sense it's affecting Canadian trade negatively with one country and affecting US trade negatively with you know.. every country.

On iOS you can deny an app cellular data access which accomplishes this, as long as you don't launch it on Wifi. But yes I too wish I could deny apps internet access completely.

Syncthing-fork is in the Play Store and works fine for me.

The commands aren't the special sauce, it's the analytical capabilities of the LLM to view the outputs of all those commands and correlate data or whatever. You could accomplish the same by prefilling a gigantic context window with all the logs but when the commands are presented ahead of time the LLM can "decide" which one to run based on what it needs to do.

Electronic warfare is pretty effective against drones that are using radio waves for their communication. Earlier in the war you could see a lot of drone footage that would become washed out with static as they got closer to tanks so it's much more reliable to use spools of fiber.

I use o3 with architect mode for larger changes and refactors in a project. It seems very suited to the two-pass system where the (more expensive) "reasoning" LLM tells the secondary LLM all the changes.

For most of the day I use Gemini Pro 2.5 in non-architect mode (or Sonnet when Gemini is too slow) and never really run into the issue of it making the wrong changes.

I suspect the biggest trick I know is being completely on top of the context for the LLM. I am frequently using /reset after a change and re-adding only relevant files, or allowing it to suggest relevant files using the repo-map. After each successful change if I'm working on a different area of the app I then /reset. This also purges the current chat history so the LLM doesn't have all kinds of unrelated context.

Aider has a configuration for each supported LLM to define the best diff format for each; so for certain ones they're best at diff format, Gemini is best at a fenced-diff format, Qwen3 is best at whole file editing, etc. Aider itself examines the diff and re-runs the request when the request when the response doesn't adhere to the corresponding diff format.

Edit: Also the Aider leaderboards show the success rate for diff adherence separately, it's quite useful [1]

1 - https://aider.chat/docs/leaderboards/

Yeah I should have clarified that, there are more advanced IDE integrations now but I meant Cursor, Copilot etc.

Using Aider with o3 in architect mode, with Gemini or with Sonnet (in that order) is light years ahead of any of the IDE AI integrations. I highly recommend anyone who's interested in AI coding to use Aider with paid models. It is a night and day difference.

The new AMD chips in the Framework laptops would be a good candidate and I think you can get 96GB RAM in them. Also if the LLM software is idle (like llama.cpp or ollama) there is negligible extra power consumption.

I considered being polite but the fact OP had already chosen to not answer these actual questions makes me question why anyone would trust this site with their payment data or even personal data.

This seems super amateur and your privacy policy sucks compared to OpenRouter. Also it's weird you have time to respond to trivial questions but not questions like "why would I use this over the entrenched leader in this space".

No docs or info without signing up, confusing Grok and Groq, and claiming you have access to o4 models which haven't been released makes this look like an incredibly unserious offering.

We (Canada) literally signed the United States-Mexico-Canada Agreement for trade in 2020... with Donald Trump. Trump this February said "Who would ever sign a thing like this?" and proceeded to make up the story about fentanyl coming from Canada as a pretense to ignore this deal. So there was certainly more than a pinky promise.

Additionally your senate just signed law to repeal the tariffs against Canada specifically so I don't know what other receipts or paperwork you'd like here. Not that I expect Trump to adhere to that of course, but I've never heard the perspective of "well these weren't signed agreements" for things that clearly were.

Could you clarify for me please why the term globalist would be used instead of global? Like "Apple is a global corporation" is an unarguable fact but using the term globalist for these companies seems pretty strange to me since there are some pretty weird connotations with it.

You know the craziest thing about watching this as an outsider in Canada is that we as a nation are losing one reliable trading partner (and obviously losing an ally but that's another conversation). Germany - losing a trade partner. Japan? And so on. But the US has lost 180 reliable trading partners. You're already seeing in the first week burgeoning trade alliances between nations who wouldn't have cooperated before, and the amount of resentment against America from this is going to take a generation to heal.

Of course I'm sure some people would claim this is 4D chess or what not but.. sometimes the person doing something crazy is just crazy at the end of the day.

Please no, all due props to Valve but they really can't count to 3.