frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Inkling: Our Open-Weights Model

https://thinkingmachines.ai/news/introducing-inkling/
82•vimarsh6739•1h ago

Comments

ls_stats•55m ago
America needs its own DeepSeek or Z.ai, a lot of people (myself included) root for open chinese models to win because they have no other choice.

Thinking Machines might be it.

verdverm•45m ago
Its not as good as GLM 5.2 for agentic workflows while also being bigger. Competition is going to be ruthless because the super low cost to switching.

There is also AllenAi in the US, but they have yet to produce a model at this scale. Thankfully, new contenders can come out of nowhere and do well, as long as they can produce a competitive model.

gkapur•44m ago
It could be but there are a host of companies going after open weights models: Arcee, Reflection, Llama (TBD on Meta's focus on closed-source versus open-source), etc.

That said, the fine-tuning API + open weight model at least is a semblance of a viable business that could work so I will be curious about it. I'm not sure the synergy is fully there (why is someone with an open weights model privelaged to fine-tune it better if it's just QLora or Lora) but let's see!

alansaber•51m ago
I never thought i'd see the day they released a model, rather than a blog post. The Figure 3 demo being a screencap of chrome in localhost made me feel better about myself. Jokes aside, best western open weights model- very cool.
pr337h4m•28m ago
They are one of the few labs (perhaps even the only one at this level) that are doing something both unique and useful, rather than simply imitating what the others are doing: https://thinkingmachines.ai/blog/interaction-models/
verdverm•48m ago
If it's ~30% bigger and not as good as GLM 5.2, why would I tinker with this model?

Maybe for the multi modal?

gkapur•40m ago
If they have a really seamless fine-tuning experience and maybe can help you extract the data you need to FT (which is one of the big challenges in actually getting fine-tuning democratized), maybe you would use it because "Tinker" defaults to it.

The model could also be more flexible for non-coding use-cases (they show the results for reasoning being strong) so maybe the argument is to use it for non-coding use-cases to drive relatively deterministic conclusions for non-coding agents (they have also done some determinism work on kernels, which could be useful in pulling on that thread of deterministic models that are fine-tuned for everything that is not writing code.)

That said, I'm not sure how much all the work they have done actually synergizes or if the market size (at least in the short to medium term) is big enough for a huge outcome from the company's current valuation with those bets as the enterprise agent estate is taking a while to evolve. Hence companies like Anthropic and OpenAI are throwing tons of consulting money at the problem.

Aurornis•33m ago
> If it's ~30% bigger and not as good as GLM 5.2, why would I tinker with this model?

The benchmarks never tell the full story. Some of the open weights models have been benchmaxxed for a while. Their utility on real work can be different than the benchmark number.

The multimodal input is also a big deal. Having vision input is really helpful for a lot of tasks.

buremba•27m ago
Then why are they publishing the benchmarks which makes them look worse than GLM 5.2?
Reubend•45m ago
Seems like this is particularly good at instruction following, but not as strong at coding as others. It's always great to get more diversity of open weight models though! I'll need to test this out to see what its "personality" is like.
ianbutler•37m ago
It's nice to see a strong long context open weights model that is multi-modal.

There are many applications that will benefit from the strength in audio here and until z.ai and co work in visual this could be very strong for general agentic applications, though I see there's a bit of weakness in the benches for areas that might make that less true.

Like all models need to slap it in your harness and do proper evals on the tasks you care about.

0xbadcafebee•12m ago
[delayed]
amarble•31m ago
They also indicate they have a 276B A12B version, but it doesn't seem the weights are available. This might actually be able to fit in 128GB when quantized to 2 bits or so which makes it interesting.
Flux159•30m ago
They mention in the announcement link https://thinkingmachines.ai/news/introducing-inkling/ that they are still testing Inkling-Small and it will also still be multimodal. This makes it super interesting as a Deepseek V4 Flash replacement (and would be interesting with DwarfStar / ds4 if it gets supported).
pants2•31m ago
The Artifical Analysis has a link on their homepage but it 404's :/

https://artificialanalysis.ai/models/inkling

raverbashing•30m ago
Cool, now we just need the GPU that supports it
janalsncm•28m ago
For the most part it’s better than Nemotron, worse than GLM. This makes it the best American open weights model from what I can tell?
nickludlam•17m ago
It's nearly double the size of Nemotron 3 Ultra, so I'd expect it to be considerably better, although the active parameter count seems to be a touch lower at 41B vs 55B
bobkb•26m ago
Happy to see an open weight model ! This has all the right ingredients for success.
solomatov•24m ago
It looks like HuggingFace shows Apache-2.0 but they have AUP. How does it work together?
firasd•20m ago
Looks like it can be tried at https://tinker.thinkingmachines.ai/playground
bbstats•19m ago
too bad we'll never know how good it is, since they used a radar plot to show its benchmark scores!
inkvi•12m ago
Do they have an api to try the model in real envs?
dr_dshiv•10m ago
What are the different business models for open-weight AI companies?
firasd•5m ago
Just serving the model over API seems like a natural fit and is what many of them are doing. So simply being the cloud provider for your own open weight model can be a source of revenue
MaxPock•9m ago
Raised 2 billion dollars at a 12 billion valuation and debuts at 41 on the Artificial Analysis Intelligence Index, while KIMI and DeepSeek will release Fable-class models this week. What a joke.
verdverm•24m ago
being close is still impressive, especially for their first (released) model

gives me hope that the training moat is even smaller than we thought

speedping•26m ago
I second that. Gemini 3.5 Flash rocks the benchmark charts but is terrible as an agent. Horrible instruction adherence and makes WAY too many tool calls
luckydata•14m ago
which cheap models have you found work best as agents?
Flux159•31m ago
There's also an Inkling-Small that is 276B, 12B active that is much smaller than GLM 5.2 and still multimodal. Not released yet, but in the announcement link they mention that they're testing Inkling-Small & will release as open weight after testing. That one may be interesting as a Deepseek V4 Flash replacement.
pizlonator•7m ago
> Maybe for the multi modal?

Yeah

Linus Torvalds tells AI haters to fork off

https://www.theregister.com/ai-and-ml/2026/07/15/linus-torvalds-tells-ai-haters-to-fork-off/5271894
1•amrrs•27s ago•0 comments

My thoughts on the future of Go in the AI era

https://packagemain.tech/p/my-thoughts-on-the-future-of-go-in-ai-era
1•der_gopher•1m ago•0 comments

Rise and Fall of Silicon Graphics,the Giant That Built Hollywood's Biggest Films [video]

https://www.youtube.com/watch?v=zE86S5n7gk8
1•oldnetguy•2m ago•0 comments

Programming Thinking [video]

https://www.youtube.com/watch?v=KtBefDeECVU
1•erichocean•3m ago•0 comments

State of CLI Coding Agents, Mid-2026

https://blog.arcbjorn.com/state-of-cli-coding-agents-2026
1•gmays•5m ago•0 comments

Reflect Open – a new chapter for Reflect

https://reflect.app/blog/reflect-open
1•allenleee•5m ago•0 comments

Iran Messaged Vance Warning That US Envoys Were "Abusing" Their Inside Access

https://www.dropsitenews.com/p/iran-jd-vance-kushner-witkoff-exploiting-negotiations
1•enraged_camel•10m ago•0 comments

ChatGPT's Wokeness Could Cost OpenAI Millions [video]

https://www.youtube.com/watch?v=RUgF7MoUetk
1•moebrowne•10m ago•0 comments

Lab 6

https://lab6.com
1•figbert•10m ago•0 comments

The Alignment Sciences Academy

https://alignment-sciences-academy.citizen-of-earth.chatgpt.site/
1•CitizenOfEarth•13m ago•0 comments

PolicyBot–open source GitHub Action enforce commit convention and PR quality

https://github.com/mogwi1983/policybot
1•jamesrodriguez6•16m ago•0 comments

Redis slashed 25% of its Tel Aviv workforce (and 200 worldwide)

https://www.timesofisrael.com/israeli-founded-data-tech-unicorn-set-to-slash-25-of-its-tel-aviv-w...
2•ohjeez•17m ago•0 comments

Tell HN: Fable Credits Are 150x More Expensive

4•yesitcan•19m ago•2 comments

Falsehoods Programmers Believe About Names (2010)

https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/
1•FigurativeVoid•19m ago•0 comments

Tipping Points: When the Climate System Flips

https://www.tobiasreithmeier.de/en/blog/ipcc-tipping-points-climate-system
1•JuriKeller•21m ago•0 comments

Five arrested after Hong Kong police raid independent bookshops

https://www.bbc.com/news/articles/cq61660qpdpo
2•tartoran•23m ago•0 comments

Three Ways to Write a Table in Podlite

https://www.perl.com/article/three-ways-to-write-a-table-in-podlite/
2•oalders•24m ago•0 comments

A Case Against SDKs

https://twitter.com/alvinsng/status/2077114275412512868
4•favoboa•28m ago•1 comments

Leaked Suno code lists 2,013,545 YouTube Music clips

https://www.theverge.com/ai-artificial-intelligence/966072/suno-ai-music-training-scraping-youtub...
1•logickkk1•30m ago•1 comments

Musk wants to open source X

https://twitter.com/elonmusk/status/2077361679034118271
2•miohtama•30m ago•1 comments

Show HN: Limits, an on-device iOS app for tracking AI usage limits

https://getlimits.app
1•treexs•31m ago•0 comments

Fourier pixels for bidirectional light control

https://www.nature.com/articles/s41586-026-10681-7
2•gnabgib•31m ago•0 comments

Every successful GPT-Red attack becomes defender training data

https://twitter.com/OpenAI/status/2077446721161093124
2•logickkk1•31m ago•0 comments

CVE-2026-15409: SonicWall SMA 1000 SSRF Zero-Day

https://hellorecon.com/blog/cve-2026-15409
1•slvnx•31m ago•0 comments

Making 768 servers look like 1

https://planetscale.com/blog/making-768-servers-look-like-1
3•aarvin_roshin•32m ago•0 comments

Dev Log #0: A look inside GSV

https://gsv.space/blog/dev-log-0/
1•deathbyknowledg•32m ago•0 comments

Show HN: Health.md, an Apple Health exporter for files and self-hosted APIs

https://github.com/CodyBontecou/health-md
1•codybontecou•32m ago•0 comments

Egyptian Afterlife Walkthrough

https://zhenyi.gibber.blog/egyptian-afterlife-walkthrough
1•zhenyi•33m ago•0 comments

Human intelligence may lie in the power of a single brain cell

https://medicalxpress.com/news/2026-07-secret-human-intelligence-power-brain.html
3•gmays•34m ago•0 comments

The Chip Is a $15,000 25-MPH EV That Makes Driving Optional

https://insideevs.com/news/801772/chip-motors-ev-remote-driving-2026/
1•atlasunshrugged•35m ago•1 comments