CJ 🇺🇸 retweeted
First Tencent Cloud. Now Alibaba Cloud.
@chutes_ai is serving GLM-5.3, DeepSeek V4.1 Flash and Qwen3.8 Max. GLM and Flash are 35% off until Nov 23.
Two of China's biggest clouds plugging into a Bittensor subnet.
Decentralised inference is going mainstream. #bittensor
After Tencent Cloud, Bittensor subnet @chutes_ai has partnered with @alibaba_cloud.
GLM-5.3 and DeepSeek V4.1 Flash are 35% off on Chutes through November 23.
CJ 🇺🇸 retweeted
Drop your token cost by 35% this week in 10 seconds.
If you run GLM-5.2 on Chutes, switching to GLM-5.3 is a one-line change: zai-org/GLM-5.3.
Z.ai reports gains over 5.2 on repository-level coding. Compare it on your own prompts while it’s 35% off, through Nov 23.
Model page and API details: chutes.ai/app/chute/chutes-z…
CJ 🇺🇸 retweeted
GLM-5.3 and DeepSeek V4.1 Flash are 35% off on Chutes until Nov 23.
GLM-5.3 is $0.80 per million input tokens and $2.50 per million output. Flash is $0.195 and $0.78.
Your agent can use Flash for everyday calls and GLM-5.3 for the hard ones, on one key. Qwen3.8 Max is on it too, at its usual price.
CJ 🇺🇸 retweeted
We've partnered with @alibaba_cloud to offer GLM-5.3, DeepSeek V4.1 Flash and Qwen3.8 Max!.
GLM-5.3 and Deepseek V4.1 are 35% off through Nov 23
Your existing key and credits work. Just change the model ID
CJ 🇺🇸 retweeted
Replying to @WSJ @WSJopinion
You sort of set him up, didn’t you, by creating the impression that senior government officials anonymously provided information about spying, when in fact the info came from a lobbyist at a competing AI company?
CJ 🇺🇸 retweeted
Do not get distracted by the headlines.
Open-source AI is essential to the freedom of billions of people in the future.
CJ 🇺🇸 retweeted
Parallax runs at almost 18,000 tokens a second on a single RTX 5090.
That’s one consumer GPU answering 256 requests at once, on a vLLM fork that we put together.
The more one card can serve, the cheaper each answer gets 💪
CJ 🇺🇸 retweeted
GLM 5.2 Fast is live on Chutes.
Z.ai's GLM 5.2, for chat and agents where the reply is on the clock. 1M-token context, 128K max output.
Here's what GLM 5.2 is good at, on Z.ai's published scores: 81.0 on Terminal-Bench 2.1, four points behind Opus 4.8 at 85.0, and 62.1 on SWE-bench Pro, up from GLM 5.1's 58.4. Z.ai built it for long-horizon tasks.
Base URL llm.chutes.ai/v1, model zai-org/GLM-5.2-Fast, same key as every other model ofc.
$2.45 per million input tokens, $7.70 out.
CJ 🇺🇸 retweeted
Bittensor subnet @chutes_ai has partnered with Tencent Cloud.
Chutes is a launch partner for @TencentHunyuan’s new flagship AI model, Hy4, now available through its API.
Day-one access to leading AI models is business as usual on Bittensor.
Tencent Hy4 is live on Chutes
We’re a launch partner with @TencentHunyuan.
It’s Tencent’s new flagship, and the first Hy model that can see. It reads screenshots and charts, and it can drive a terminal or a desktop app for you.
0.525$ per million tokens in, 1.5625$ out.
Already on Chutes? Just Change the model name to start using it
Try now: chutes.ai/app/chute/chutes-t…
CJ 🇺🇸 retweeted
Tencent Hy4 is live on Chutes
We’re a launch partner with @TencentHunyuan.
It’s Tencent’s new flagship, and the first Hy model that can see. It reads screenshots and charts, and it can drive a terminal or a desktop app for you.
0.525$ per million tokens in, 1.5625$ out.
Already on Chutes? Just Change the model name to start using it
Try now: chutes.ai/app/chute/chutes-t…
CJ 🇺🇸 retweeted
The midterm results, regardless of your opinion on them, come with a fairly high degree of uncertainly around the future of new datacenter capacity (and/or costs) in the US. Parallax is no longer really optional IMO (though it was already inevitable).
“You never change anything by fighting it; you change things by making them obsolete through superior technology.”
- R. Buckminster Fuller
Status updates:
Almost done with dataset prep. It turns out ~10t tokens of high quality data tokenized and properly upsampled and so on is a lot of work and compute.
Apple M5 and DGX spark support locked in, still waiting on my AMD Ryzen AI MAX+ PRO 495 from lucebox to get that supported as well.
Still chugging away at the final touches on the code/agentic pipeline also (but that would be after the pretraining anyways, non-blocking).
Settled on the openai o200k tokenizer, updated to always use single tokens for digits for better math capabilities (hopefully).
(and more to come soon)
CJ 🇺🇸 retweeted
DeepSeek V4.1 Flash is live on Chutes.
Pair it with V4 Pro: route the routine calls to Flash and the hard ones to Pro. Same key, different model id.
Flash is the lighter of the two: 552B parameters, with 8B active per token while it reads and 16B while it writes, against Pro's 1.6T. With the same 1M-token context window
On agent work it holds its own. On DeepSWE v1.1, DeepSeek's published score is 77.0, ahead of Opus 5 at 74.0, and on Agents' Last Exam it tops DeepSeek's table at 35.4. It trails Opus 5 on Terminal-Bench 4.0 (42.3 against 51.8) though, so keep Pro for the longest terminal sessions
$0.30 per million input tokens, $1.20 out.
V4 Pro is $0.85 and $2.65.
CJ 🇺🇸 retweeted
One of the best Open LLM for us.
GLM 5.3 is live on Chutes.
It's Z.ai's flagship, built on the same base as GLM 5.2 and retrained for complex coding and long agent runs. 1M-token context, open weights.
Here's what to use it for:
- security review: on CyberGym, Z.ai's published score is 84.5, ahead of Fable 5 at 83.8 and Opus 4.8 at 78.1
- coding agents in a terminal: 88.2 on Terminal-Bench 2.1, ahead of Opus 4.8 at 85.0
Over GLM 5.2: 81.0 to 88.2 on Terminal-Bench 2.1, 46.2 to 66.9 on DeepSWE v1.1.
It does trail from Fable 5 and GPT-5.6 Sol on the longest software-engineering runs (66.9 on DeepSWE v1.1 against 69.7 and 72.7) though, so test it on your own repo first.
$1.225 per million input tokens, $3.85 out.
Z.ai's own API lists $1.40 and $4.40.
CJ 🇺🇸 retweeted
Kimi K2.7 Code is live on Chutes.
Base URL llm.chutes.ai/v1
Model moonshotai/Kimi-K2.7-Code
Works with any OpenAI-compatible client, same Chutes key.
Moonshot built it on K2.6 for coding agents: 1T parameters with 32B active per token, a 256K context and open weights. Moonshot says it uses about 30 percent fewer thinking tokens than K2.6, so a job takes fewer tokens.. making it more efficient
$0.83 per million input tokens, $3.50 out.
If you’re looking for a solid coding model at a reasonable price then you should give this one a shot.
CJ 🇺🇸 retweeted
Cheap inference gets expensive when it fails.
@OpenRouter has begun routing more traffic toward fast, reliable inference pushing providers to compete on quality.
As a provider, we welcome the competition.
We just shipped our biggest improvement to our provider router since the founding of OpenRouter.
Before, we prioritized mostly based on price. The new router now rewards the best quality inference for the actual cost of a request, so requests stay cheap but are also fast and reliable.
Providers that deliver more value automatically win more traffic, so every provider has a reason to keep improving. And every time a provider gets better, so does your inference on OpenRouter.
CJ 🇺🇸 retweeted
An open model now sits a tenth of a point behind Fable 5 at coding in a terminal.
That's DeepSeek V4 Pro on Terminal-Bench 2.1: 87.9 against Fable 5's 88.0, with Opus 4.8 at 85.0 (DeepSeek's published scores).
What to use it for:
- a coding agent that runs commands, reads the errors and fixes what broke
- security review, where it tops DeepSeek's Cybergym table at 83.3
Its 1M-token context lets you hand it a big slice of your repo at once, so the agent sees how the pieces connect instead of guessing file by file.
It's weaker on long, multi-step projects (62.7 on DeepSWE against Fable 5's 70.0), so try it on your own tasks before trusting it with those.
On Chutes it's $0.85 in and $2.65 out per million tokens, at any hour.
Model page and pricing: chutes.ai/app/chute/chutes-d…
Much of SaaS is reselling inference as part of its platform.
That business line lives on the gross margin after the token bill, and labs and clouds are selling tokens at a discount to win share.
It's a debt/VC funded offering at this point. Everyone in SaaS will need to answer this question.
release all the numbers.
does 80% yoy become 65% when you remove passthru inference?
what does it become when you remove ai natives?
and how much of it is from one time software license --> service transition?
fcf, RPO, cRPO?
op margin non-gaap, gaap? dilution?
curious ppls want to know
ps: @alighodsi happy for your success. the world is big and life long.
nitter.cf/bhorowitz/status/21150…
1
126
CJ 🇺🇸 retweeted
DeepSeek V4 Pro is live on Chutes.
It's the big one from DeepSeek's V4 family, with 1.6T parameters and a 1M-token context window. Open weights, MIT licence.
It's strong on agent work. DeepSeek's published score on Terminal-Bench 2.1 is 87.9, ahead of Opus 4.8 at 85.0.
$0.85 per million input tokens, $2.65 out.
Try it here: chutes.ai/app/chute/chutes-d…
CJ 🇺🇸 retweeted
Chutes is very excited to be working with METANOVA LABS as they develop their Hit-to-Lead optimization agent! We fully intend to assist and support to this project in any way we can as it develops.
NOVA Hit-to-Lead Optimization Agent
In collaboration with @chutes_ai and @onepot_ai
We're developing a Hit-to-Lead optimization agent that integrates an algorithm developed by a NOVA Blueprint winner, extending NOVA's capabilities beyond hit discovery and into the iterative optimization of small molecules.
The system combines a domain-specialized AI agent with proprietary combinatorial chemical databases supplied by our partner Onepot AI, spanning billions of possible compounds, that are delivered within 5-7 business days.
Researchers will be able to guide the agent in developing, comparing and refining optimization strategies, exploring chemical modifications and prioritizing candidates based on predicted affinity, selectivity, physicochemical properties and specific binding interactions.
Confidentiality is a central design requirement. Pharmaceutical R&D involves valuable intellectual property, including compound structures, targets, emerging chemical series and optimization strategies that cannot be exposed publicly.
The Chutes team integrated domain specific models so that we can run inference in their verified Trusted Execution Environments (TEEs), with encrypted communication and controlled access to proprietary databases and project records. This architecture is intended to support confidential inference on external infrastructure while preserving organizational control over sensitive research data.
Finding a molecule with promising predicted affinity is only the beginning. Hit-to-Lead optimization involves balancing multiple, often competing objectives, including potency, selectivity, physicochemical properties and synthesizability, before progressing toward experimental activity and ADME validation.
The next step is to integrate the agent into our own Hit-to-Lead development program.
For NOVA, this represents an important step toward translating algorithms developed through open competition into tools for proprietary drug discovery programs, including our own asset-development pipeline and those of external research partners.
Confidential open-weight models from @chutes_ai sit right next to claude, gpt & gemini on SayGM.
Same key, same code.
Your prompt stays sealed twice: once in our gateway TEE, again in chutes’ attested hardware.
SN28 x SN64 🤝
saygm.com/blog/access-chutes…

























