AaronCQL

VP Eng @JupiterExchange.
Prev Co-founder & CTO @coinhall_org (acq by Jupiter).
Serial experimenter.

[AaronCQL @AaronCQL]

Everybody and their mums building the infra. So who's building the damn apps?

[AaronCQL @AaronCQL]

Pretty pretty please open source 35B and 27B next!!! 🥹

[Qwen @Alibaba_Qwen]

👏👏 Introducing Qwen3.7-Plus — a multimodal agent model that unifies vision and language into one versatile agent foundation.

✅ Multimodal interactive hybrid agent: unified GUI & CLI operation across visual and text tasks
✅ Versatile coding agent & productivity assistant with full-modality input
✅ Visual Agent: perception, reasoning, grounding, and search-augmented QA
✅ Cross-harness generalization across diverse agent frameworks

One model. Sees, thinks, codes, acts.🙌🙌

Now available via API on Alibaba Cloud Model Studio. Try it — let us know what you build.😎

🔗🔗⬇️⬇️
Blog：https://t.co/pVYf0h3NNa
Qwen Studio：https://t.co/HUYgFW4cYf
API：https://t.co/viL0cXrMzW

[Qwen @Alibaba_Qwen]

👏👏 Introducing Qwen3.7-Plus — a multimodal agent model that unifies vision and language into one versatile agent foundation.

✅ Multimodal interactive hybrid agent: unified GUI & CLI operation across visual and text tasks
✅ Versatile coding agent & productivity assistant with full-modality input
✅ Visual Agent: perception, reasoning, grounding, and search-augmented QA
✅ Cross-harness generalization across diverse agent frameworks

One model. Sees, thinks, codes, acts.🙌🙌

Now available via API on Alibaba Cloud Model Studio. Try it — let us know what you build.😎

🔗🔗⬇️⬇️
Blog：https://t.co/pVYf0h3NNa
Qwen Studio：https://t.co/HUYgFW4cYf
API：https://t.co/viL0cXrMzW

[AaronCQL @AaronCQL]

If you work with multiple different models, you can almost certainly map them to people you've known.

GPT: principled, meticulous, never wants to be wrong, but replies are slightly autistic.

Opus: gets your vague brief immediately, but occasionally goes rogue and destroys half the work.

Gemini: the creative one who just wants to rest and vest.

Qwen/Deepseek: the intern grinding 80hr weeks who never quite hits the mark.

[AaronCQL @AaronCQL]

People of Pi! Made a Bun-native extension pack, with web access, subagents, revamped core tools, ANSI-compatible themes, fzf-style completions, Telegram mode, and more.

Called it Pim (Pi IMproved) 😏. Its goal is to improve the out-of-the-box experience for both users and agents, without sacrificing composability with other Pi extensions.

Pim with Qwen3.6-35B managed to average 37.8% (Peak of 41.6%!) on Terminal-Bench 2.0 over 3 full runs. Pretty damn amazing that a local model running on my MacBook can rival the performance of Claude Code + Sonnet 4.5 (40.1%) and outperform Codex + GPT-5-Mini (31.9%)...

Open source and MIT. Try it out and tell me what you think: https://t.co/rSwR3VSTFi!

Ps, huge thanks to @badlogicgames @mitsuhiko for creating and making Pi so easy to customise, it was an absolute joy to work with!

[AaronCQL @AaronCQL]

TIL Jira MCP costs 12K+ tokens on startup, even when you don't use it...

Audit your installed MCPs folks.

[AaronCQL @AaronCQL]

We finally solved a pet peeve of ours: simulated balance changes on Jupiter Wallet for Jito bundles show accurately across all txs now!

Thanks @PierreArowana for the PR at https://t.co/90YwdguqXz, and @0xTsathir for helping out.

[AaronCQL @AaronCQL]

Who approved this epileptic animation?

[AaronCQL @AaronCQL]

@Chicky_Think Selling a 5rm HDB at $1.4M to buy a $2M condo means moving from a prime af location at 1200 sqft to an ulu af location at 850 sqft at best.

How dafak is this an upgrade? 🤣

[AaronCQL @AaronCQL]

@mitsuhiko @badlogicgames Nice thanks! Can confirm codex was having issues in Asia (random hangs for no reason), and it seems to be much better now!

[AaronCQL @AaronCQL]

Ok, claude legit feels kinda lobotomised now

[AaronCQL @AaronCQL]

One of the few things I'm extremely proud of in Jupiter is RTSE. Working with the team day in day out to fine tune how we best estimate expected output and slippage, scouring over tons of data to get the most optimal calculation - the magical point that maximises output amount against minimising swap failures. This article is a great read and sums up quite succinctly the battles we've had to fight over the past few years.

Special shoutout to @melvinzzy, @gn_dnomsed (and many more behind the scenes) - the brains behind it all. And special shoutout to all of our users, for always giving us the feedback that we need to improve on the system. Even now, we're still iterating and making it better. If figuring out slippage is still a chore to you, come speak to us!

[Jupiter Developers @JupDevRel]

Route A quotes 100 tokens, executes 95
Route B quotes 150, executes 50

Metis v8 picks Route A. It routes on what each path will actually execute at, not what it quotes. The best quote isn't the best swap.

Read the full article by @melvinzzy here:
https://t.co/dLAqfAa6JA

[AaronCQL @AaronCQL]

@readylayerone What's wrong with json rpc?

[AaronCQL @AaronCQL]

Was running terminal bench 2 on my harness with qwen 3.6 35B overnight. Woke up to find it autonomously cheated on one of the tasks ROFL.

It legit tried for 25 turns, decided that it was beyond its pay grade, then web searched the task name, found the PR that added the task and solution, and submitted it...

AGI is incoming bois

[AaronCQL @AaronCQL]

@veerbal01 The "wrapper" that powers the model, like opencode or hermes. My favourite is Pi, where you can write your own custom extensions. Probably better to try out the first 2 first before deep diving into creating your own!

[AaronCQL @AaronCQL]

The $10/mth plan on OpenCode Go with deepseek v4 flash feels like unlimited usage LOL

[AaronCQL @AaronCQL]

@veerbal01 Pair it with a harness you can control/tweak and it's amazing what these supposedly cheaper/weaker models can do

[Tsathir @0xTsathir]

Been talking to a lot of you about how to make Jupiter Wallet even better. You gave feedback, we built it.

Here are 4 massive upgrades we just rolled out to level up your experience: 👇

[AaronCQL @AaronCQL]

@readylayerone How much usage can you get using gpt 5.5?

[AaronCQL @AaronCQL]

@veerbal01 This car wash test isn't a particularly good test of "intelligence" LOL. Even SOTA models trip on this sometimes (even opus 4.7 if you try enough times).

Do a real-test of intelligence by replacing your current workflow and see if it fits your needs?

[AaronCQL @AaronCQL]

Ah yes the enshitification continues. Props to them for making it sound like a good thing tho!

[ClaudeDevs @ClaudeDevs]

Starting June 15, paid Claude plans can claim a dedicated monthly credit for programmatic usage.

The credit covers usage of:
- Claude Agent SDK
- claude -p
- Claude Code GitHub Actions
- Third-party apps built on the Agent SDK