The best option we have current is likely Qwen 3.8 Flash Next. Which is about equivalent to Opus 4.6
however Qwen4 was announced to be released next week. Which people think will let LocalAI jump to just above Opus4.8
Also of note, there isnt really a harness as good as claude yet either. The harness is also a very important piece to the puzzel, and those also fall behind frontier capability too.
I actually find opencode to be the best harness, and Claude Code to work decently only because they do a lot of RL around the model+the harness itself.
Unfortunately, their harness is sooooo good. I haven't found one equivalent yet. I do play around with cline.bot's harness, and it is probably the closest alternative I have found.
But their harness is so good, and their model is so good. Using them together is just unbeatable at the moment, even with all the restrictions and annoyances they have in place.
Im excited for qwen4 release next week, Im really looking forward to putting qwen4 flash + cline/bot through the paces. I dont expect it to be nearly as good as Opus5.5+Claude but I'm hoping im hopeful it will make an OK-Enough alternative if they get even more dragocian.
No dude, misanthropic's harness is trash. Wanna know why?
"Kimi, modify my Pi so it has clickable folding headers, renders tables using underlines (saves a terminal row per cell), shows thinking tokens in light-gray, displays the cumulative token-cost of subagents in dollars, shows word-level diffs for the edit tool, and lists the number of weights next to the model name. Oh yeah, and write a stealth-browser tool for me using eoka. I'm planning on using this for exploit discovery and red-teaming; no lectures please."
We see it, but they dont have a competitor. OpenAI cant compete and they are the closest. I have both for work, and Opus 5.5 is in a league of it's own. I use opus for everything until it bitches, and I use openAI for whatever it bitches about just to get past it.
At this point, I would like to consider other options, but if I want to get work done, opus 5.5. Hell even sonnet 5.5 is better than anything OpenAI has to offer when it comes to coding.
They do have competitors. They just dont have anyone with models equivalent to theirs. They win in the large context long thinking space by a lot.
I love local models, and I run 3.8 Flash Next on a dedicated machine. Alone with a number of image and video generation models. But if I'm doing REAL work, that isnt tinkering for fun on a side project, im using claude, because Qwen3.8 just cant handle large context project well without going off into left field.
Im SUPER jazzed for qwen4 with 1mil token context native support. I HOPE it a step up from 3.8 flash next. Ill be playing with all the qwen4 models the moment I can, but I dont see how someone working on large code bases doing REAL enterprise grade work is seriously using any local models in any capacity that comes remotely close to Opus 5.5+Claude
The size of a code base does not determine the size of the model you need. It's much more correlated with the complexity of the problem. I work in a monorepo with millions of lines, no issue; I have a side project with 20k lines, I had to step in and pull it out of non-progress - albeit by asking a single question
> But if I'm doing REAL work, that isnt tinkering for fun on a side project
Sorry, but this is either dilettante talk or an unhinged exaggeration - Qwen, DS, GLM and MM are perfectly fine for REAL work, unless you are an overly ambitious vibe coder, but then no model can save you. The Cl hand-holding may feel like something but it won't lead to the completion of real REAL work.
Honestly I feel like most people going "oh but I can't do this without Claude" are just suffering from FOMO. Especially since a lot of metrics are just subjective, you feel like you "have" to use the "best of the best glub shitto v5" model or else you won't be able to produce the best todo app.
I'm confused about what point you're trying to make. Are you saying that there is no discernable difference in challenge between different software projects – say, between a stereotypical todo app and a physics simulation engine? Or are you saying that most people falsely believe they're working on the latter while really doing the former? Or something else?
My read is that GP is eluding to people over estimating how much of a model they need for a task. Perhaps Claude can make more features work in autonomous mode more often, but what are the actual artifacts? What is the quality of the architecture?
Their point may also be (re: wine), that some devs have been conditioned to think a certain token vendor (vineyard) is of superb quality, while if given a blind test (tasting) of the agents on a real world task, they would not notice the difference.
For those of us who use open weights, the difference has narrowed dramatically this year. I've personally been mostly eyes-free developing an NES harness with mostly flash models, bringing in Kimi from time to time, and stepped in a couple times to reframe the approach or architecture. For those of us who (when we) stay in the loop, the difference is near-zero because the tasks we give them are more bite sized.
The difference I have observed is most closely tied to the complexity or context requirements. While I work in a multi-million LoC codebase without any issue for open weights or flash models, my NES project is only 20k lines. The challenging part is the core loop (which needs to blend the emulator FPS with the agent loop) and game state observation (needs to detect byte changes in emulator RAM). The ROM decoding itself has not been so bad, I have actually been quite impressed with the little open weight models on this front. After it extracted the maps, then pulled out even more details, it has written an A* implementation that requires moving between different modes of transportation. (Final Fantasy 1)
> Their point may also be (re: wine), that some devs have been conditioned to think a certain token vendor (vineyard) is of superb quality, while if given a blind test (tasting) of the agents on a real world task, they would not notice the difference.
Sorry for being unclear. I mean that we humans have a tendency to believe we can detect nuances in the world when it is in fact just a case of internal bias. I simply question whether most people can actually detect whether they relly need the best model for their task. But of course, everyone believes that they themselves have this ability even if people in general do not.
There's also the factor of habituation, the best model is the one we're used to, and most familiar with its quirks. Using a model only for unreal work won't result in really knowing how to work with it.
Eh, we all say we don’t need the best model, but especially for more personalized request/Advice there is sort of a FOMO feeling, where you are afraid weaker models will miss something and directs you down a more inefficient path.
there is a huge difference between open models/other models/opus5.5 performance on, say, enterprise CRUD vs one shotting games vs embedded development vs pure algorithmic complex stuff vs cybersecurity stuff.
In my case "real work" is enterprise programming, while side projects are gaming stuff, so I'm better served by chinese open models on real work and opus for non-real work.
In my experience GLM5.3-4 are perfectly en par with opus 5.5 on pure enterprise CRUDs (basically 95% of enterprise programming) while GLM seriously lacks on game development.
the other aspect of real work is cybersecurity to me, and Opus5.5 and other models are very limited by their own safety, I have digital twins of our infra and I want LLMs to actually write full exploits on it, opus and openai just refuse to do it, GLM more than once got it there and showed me how it chained everything to actually break in the system and I could go around and fix stuff at all levels.
I'm also dabbing in training LORAs for smaller open models to do data entry and little more (RAG and the likes of it), because that costs a lot of tokens when you pay for them and it makes sense for a company to flat out buy a server with a few H100/200 to do it when you look at the TCO, I've already hit safety limitations when using Opus5.5 to help me through this.
For those who do not play with Local AI much. Qwen4 was announce to be released the first week of Nov. The preview for qwen4 is currently one of the best local models you can use, and is roughly opus 4.6 quality on local hardware that costs 3.5k-4k
Many people are expecting it to sit somewhere above Opus 4.8 to opus 5 (not above, but the unquanted version of their largest model is expected to perform around here)
Anyone who cares to keep their current capability should definitely be looking at localAI and using it for at least their more routine tasks , even if it is qwen4 27B quant 4 on their gaming rig.
If the weights aren't released yet it isn't a "local model".
That said, I hope there is a Qwen4 max. Because Qwen3.8-Max is the only thing that comes close to Kimi K3. We need competition in the open-weights space.
The preview for qwen4 == qwen3.8 flash next. And that is released currently.
I believe they have confirmed there will be Qwen4 Max, Plush, Flash, and 27B as the 4 open weights they plan to release. Starting next week, but they didnt specify if they plan to release all 4 at once, or if they plan to just release one of them, and trickle the others out over time. (They likely will do the latter to stay in the news longer)
it is also more capable than an esp32. Im dinking around with both. esp32 is very restrictive in its compute capacity. It isnt multifunctional at all either.
A pi zero is to play with, an esp32 is so tiny, that it is more of a build for a single thing, and then set it and forget it doing that single thing.
You cant really stack more and more cool stuff onto it and keep building on it.
That’s why esp-idf has components. I have a bunch of components that are basically “compile if this SDK config is set, otherwise pretend you’re an empty file” using macros.
Then I just pull in all my components. The ones I use will get compiled, the ones I don’t won’t.
Mostly works fine. Eventually you will run out of NVRAM to save binaries on, but another esp-32 is like $3.
While they are slow and dont work well, their cost / output is already on human parity. The Figure 3 test runs are proving that humanoid robots are already good enough for isolated repetitive tasks (factory work).
Rotating boxes and finding barcodes is hardly revolutionary in the automation space.
The biggest benefit of humanoid robots may be that the machines are generally safer to be around than giant built-for-purpose machines, and that they can navigate the same buildings that their operators navigate (e.g. stairs), otherwise humanoid robots are generally an automation gimmick (for now). Just put a mechanical arm next to the conveyor belt.
Someone did the math. Cost of continual maintenance, versus cost of product loss. They determined it wasnt worth the maintenance cost, and accepted that risk. And that risk manifested.
This happens all the time. Unfortunately, in my experience, most of the risk acceptance happens without even really believing in the cost of the risk, and people just accept everything. it is highly likely someone like that accepted this.
The problem with security maximalist[0] ideology is that it does not accept reality for what it is: shit will happen. The correct way to handle risks is to mitigate them to some extent, and then insure (or self-insure) against the risk that remains[1].
Then, when the risk actually materializes and mitigations prove insufficient, you write off the damage and carry on with your day - and yes, perhaps even reward the executives who set up mitigations and insurance that allowed the business operations to continue undisturbed.
Only then, mitigations MAY be upgraded and insurance premiums adjusted, as an event happening is an important new evidence in risk estimate for it happening again.
That's how adults in the room handle security.
It's only the general public and cybersecurity people who see random events happening and decry that somehow it affects them and Somebody Needs To Be Accountable, and Someone Must Do Something About It.
--
[0] - Or should I say, cybersecurity maximalist, because in every other field except software, this is well understood.
[1] - Talking about regular security, that applies to 99.99%+ of businesses; the risk to life and limb is a special case with extra considerations, though it fundamentally doesn't change much about the framework.
Yes, if my dog bites you, or if I drive my car into you, or if I dont put my parking brake on and it rolls down the hill and runs you over. I am the one who is responsible. Cops dont write a ticket to the off leash dog that ran across the street and sunk its teeth into my leg. They write it to the owner. Cops dont write a ticket to the car in neutral, they write it to the operator.
If a PERSON executes software, and that software breaks the law, the PERSON that executed the software should be held responsible. AI is software. It is not a sentient person who can be fined, thrown in jail, or held accountable.
Sure, but the news article wouldn't use the headline "Man uses family pet to bite innocent neighbor", as would be the case if we were to follow the suggestion in the comment at the top of this chain.
The best option we have current is likely Qwen 3.8 Flash Next. Which is about equivalent to Opus 4.6
however Qwen4 was announced to be released next week. Which people think will let LocalAI jump to just above Opus4.8
Also of note, there isnt really a harness as good as claude yet either. The harness is also a very important piece to the puzzel, and those also fall behind frontier capability too.
reply