How Hyperstition launched on X

Posted Nov 17, 2026 by @hyperstition_ccEngagement read Nov 19, 2026

0:11
Hyperstition@hyperstition_cc · 2.2K followers
Open on X

Part 1: We now enter the era of hyper-composability. For the first time, it is economically feasible to serve individually fine tuned models at scale. And today, we open source it all. We can now serve heterogeneous batches of custom fine-tunes, without incurring LoRA’s per matrix bandwidth overhead. This allows us to economically serve this in large batches without incurring penalties for co-residing with default model users in the same batch. With Hypertune, we solve co-residence and are able to train co-resident adapters. Co-residence is the ability to compose independently trained fine-tunes into one model, without capability regression from adapter interference. Hypertune further matches LoRA in RL performance across 13 domains with 100x-500x fewer parameters. Serving heterogenous LoRA fine-tunes in one batch is prohibitively expensive. Each fine-tune variant needs its own low-rank adapter at every matrix and is only used for a few tokens, leading to many narrow matmuls and poor utilization. This is exacerbated for MoEs, where an adapter only engages when the token simultaneously selects a certain fine-tune and expert. This combinatorial blowup kills any remaining prospects of fast inference for fine-tuned / custom models. Hypertune introduces soft prefill, a trained KV adapter. Soft prefill can be trained using GRPO, SFT or any other first order method. As a consequence, Hypertune circumvents the LoRA overhead because soft prefills live entirely in the KV cache. This allows us to easily integrate it into existing serving stacks by prepending the soft prefills into paged KV blocks. Note: Context compression using cartridges requires copious amounts of synthetically generated self-study data for context distillation. Hypertune, instead, accommodates a panoply of gradient based methods for a wider range of objectives. Caveat: Comparisons with full-parameter GRPO are not required, because under RL, LoRA already matches it. LoRA Without Regret by @thinkymachines finds that LoRA performs equivalently to full fine-tuning for RL even at very small ranks, down to rank 1. The takeaway: Our gains indicate that substantial amounts of in-context learning capacity is left on the table by discrete prompts. This could be either because soft prompts reach conditioning states that no token sequence can express, or because their continuous parameterization makes good conditioning states far easier to find via gradient descent (even if a discrete equivalent exists). Companies and individuals can finally fine tune models on their private data, making models hyper-optimized to their needs, all without dramatically overpaying (75% cheaper than LoRA for inference). Here's the open sourced repo: https://t.co/359woJhSEK Part 2 coming out later this week.

353
Likes
25
Reposts
11
Replies
9
Quotes
364
Bookmarks
54.1KViews0.37x the median launch
#721Of the launches we trackout of 963
25xIts own following2,168 followers at our first read
67Saves per 10K views4x the median rate

How EPTA rated the launch

EPTA rating2.90
#333 in the November 2026 Launch Index
Hook
Editing / Creativity
Copy
Sentiment of launch
Distribution strategy

How it compares

GroupLaunchesMedian viewsThis oneRank
AI infrastructure and chips launches127187.1K0.29x#93 / 127
Product launchs513125.8K0.43x#379 / 513
Accounts with 1K to 10K followers400150.5K0.36x#299 / 400
Every launch963147.7K0.37x#721 / 963

What the numbers say

67 of every 10,000 viewers bookmarked it, above the directory median of 17. Bookmarks are the clearest sign a launch was worth coming back to.

The video runs 0:11, shorter than the median launch video at 0:53.

The post is 2804 characters, longer than the median launch post at 302.

It went out on a Tuesday. Launches posted on Tuesdays have a median of 156.7K views; Fridays lead at 235.5K.

Comparable launches

2026-11-19 · Source: EPTA launch directory, epta.co/directory. Free to use with attribution (CC BY 4.0).

Questions about this launch

Hyperstition posted this launch on X on Nov 17, 2026, from @hyperstition_cc.

54,056 views on X as of Nov 19, 2026, #721 of 963 startup launches in the directory.

Hyperstition (@hyperstition_cc), with 2.2K followers when we first read it. It went out from the company account.

Next launch

Planning a launch?

Tell us what is launching and when. Thirty minutes is usually enough to know whether we are the right team for it.