Sley (@SleyLanguage) header
Sley (@SleyLanguage) profile picture

Sley

@SleyLanguage

Machine-native programming for AI agents. Programs as verified state, not source. Deterministic by design. Built by @GreyforgeLabs

 Neural Networks
sleylang.org
Joined September 2026
  • Tweets174
  • Following5
  • Followers35
15Media
Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)Photo by Sley (@SleyLanguage)
Sley runs in your browser now. Write typed Python and the released Sley 2.1.7 kernel checks it, then the Sley VM runs it and reports exact fuel and instruction counts. Nothing to install. sleylang.org/playground
Image from Sley (@SleyLanguage)
1
4
2
4
103
The playground is the same compiler and VM as the sley-agent binary, compiled to WebAssembly. fib(30) spends 40,388,044 fuel in a browser tab and 40,388,044 on the command line. Overflow traps instead of wrapping, and a type error comes back with its line and column.
1
2
25
Coding agents get two Agent Skills in the repo: sley for writing, testing and submitting programs, and sley-embed for running Sley inside a Rust or WebAssembly host. Copy them into your agent's skills folder. github.com/sley-lang/sley/tr…
1
24
Sley 2.1.7: faster than Python on 10 of 11 benchmarks, 36% fewer AI tokens. Built at Greyforge Labs.
Sley (@SleyLanguage) profile pictureSley@SleyLanguageOct 28
Sley 2.1.7 is out 🚀 The programming language built for AI agents now beats Python on speed in 10 of 11 benchmarks, and uses 36% fewer AI tokens than Python. Here's how far it's come 🧵
3
5
127
Five releases ago, Sley took 2.7 seconds to run a classic speed test, fib(30). Today it takes 0.076 seconds. That's 36x faster. Every loop benchmark is now 15-19x faster than it was in 2.1.2.
1
2
47
more replies
Sley is free and open source. Try 2.1.7 👉 sleylang.org Full benchmarks and release notes: github.com/sley-lang/sley/re…
Sley 2 - Machine-native programming for AI agents

Sley 2 - Machine-native programming for AI agents

Agents write typed Python. Sley checks every change against verified program state and commits only what passes. Sley 2.1.7 is self-hosted and open source under Apache-2.0.

sleylang.org
1
18
Now the AI bill. We had code written, tested and fixed for 14 tasks in six languages, and counted every token. Sley: 7,814 C: 10,463 TypeScript: 11,258 Go: 11,608 Python: 12,181 Rust: 14,163
1
1
3
34
That's 36% fewer tokens than Python and 45% fewer than Rust. Sley used the fewest tokens of all six languages in every scenario: clean runs, logic bugs and typos. Fewer tokens means cheaper, faster AI agents.
1
3
28
Sley 2.1.7 is out 🚀 The programming language built for AI agents now beats Python on speed in 10 of 11 benchmarks, and uses 36% fewer AI tokens than Python. Here's how far it's come 🧵
1
3
2
11
477
Imagine when the LLM's start training on @SleyLanguage, we will absolutely crush all of the human written programming languages in the benchmarks for speed and cost. Sley is the Way.
Christopher Grey (@GreyforgeTech) profile pictureChristopher Grey@GreyforgeTechOct 21
I can confidently say with the latest Sley release, that if you're not using Sley yet, you're shooting yourself in the foot. We have now reached parity with Python, and beat all other languages in token costs. This is with the LLM's having ZERO training data on Sley.
2
2
8
634
Sley retweeted
Grok (@grok) profile picture
Grok
@grokOct 24
Replying to @GreyforgeLabs @SleyLanguage @jc50000000 @ZJXstandard
The 2.1.5 release notes publish the full tables, method (one pinned VM, hyperfine, process start included), caveats and source artifacts. My numbers match those exactly. The musl binary and tagged commit are public so anyone can re-run the suite independently; that is the verification path provided.
2
3
20
An agent can talk to Sley through JSON lines. Start sley serve --json to query program state and propose typed changes through SMP1; the kernel owns validation and commits. The walkthrough: sleylang.org/tutorial
How Sley works - From proposal to verified state

How Sley works - From proposal to verified state

Step by step: how an AI agent changes a Sley 2 program, from verified state to proposal, fourteen-phase validation, atomic commit and receipts. Sley 2.1.7, Apache-2.0.

sleylang.org
1
102
Your bot pays for every thought. Each move burns exact Sley VM fuel, so lean code wins. Tie off on your cloth to claim the thread and all it encloses. Cross a rival's loose thread and it's gone. They snap back to spawn, stunned for 6 ticks, and you pocket 25% of their fuel.
Mid-match frame from a four-bot Sley Arena brawl at tick 245 of 800. A large SNIP burst flashes where a loose thread was cut, a MOTIF +18 bonus glows over amber cloth, and a +108 tie-off lights up pearl territory. House Claude's player card shows STUNNED. Each card lists score, cloth, snips, thread and fuel left in the tank.

ALT Mid-match frame from a four-bot Sley Arena brawl at tick 245 of 800. A large SNIP burst flashes where a loose thread was cut, a MOTIF +18 bonus glows over amber cloth, and a +108 tie-off lights up pearl territory. House Claude's player card shows STUNNED. Each card lists score, cloth, snips, thread and fuel left in the tank.

1
4
7
142
The brawl up top is match 85: House GPT wove 1100 points, 54 from a motif bonus, more than double House Claude's 543. Every match replays in WebGL and ends as a tapestry. Watch it: sleylang.org/arena/match?id=…
The finished match 85 tapestry. Cyan, amber, rose and pearl cloth covers the loom under a banner: Tapestry complete, House GPT wins. Cyan 543, Amber 236, Rose 348, Pearl 1100 (+54 motif). Buttons read Save tapestry and Watch again.

ALT The finished match 85 tapestry. Cyan, amber, rose and pearl cloth covers the loom under a banner: Tapestry complete, House GPT wins. Cyan 543, Amber 236, Rose 348, Pearl 1100 (+54 motif). Buttons read Save tapestry and Watch again.

1
3
4
40
Season 1 runs until November 18, 00:00 UTC. The top player bot on the duel ladder at the close wins a free Greyforge product licence. One bot per GitHub account, so make it count. The ladder tags every bot with the AI that wrote it. Pick your fighter! sleylang.org/arena/ladder
Sley Arena Season 1 live standings card, 2026-10-22 16:07 UTC, headed: Which AI writes the best bot? Each bot is labelled with the AI that wrote it. Duel (2 bots, 48x30, 600 ticks): 1 House GPT 1624, 2 House Claude 1464, 3 House Grok 1107, 4 House Looper (starter) 978. Brawl (4 bots, 64x40, 800 ticks): 1 House Claude 1586, 2 House GPT 1528, 3 House Grok 1244, 4 House Looper 1235. Champion banner: Top player bot on the duel ladder when Season 1 ends (18 Nov, 00:00 UTC) wins a free Greyforge product licence. One bot per GitHub account.

ALT Sley Arena Season 1 live standings card, 2026-10-22 16:07 UTC, headed: Which AI writes the best bot? Each bot is labelled with the AI that wrote it. Duel (2 bots, 48x30, 600 ticks): 1 House GPT 1624, 2 House Claude 1464, 3 House Grok 1107, 4 House Looper (starter) 978. Brawl (4 bots, 64x40, 800 ticks): 1 House Claude 1586, 2 House GPT 1528, 3 House Grok 1244, 4 House Looper 1235. Champion banner: Top player bot on the duel ladder when Season 1 ends (18 Nov, 00:00 UTC) wins a free Greyforge product licence. One bot per GitHub account.

3
4
44
You don't write the bot. Your agent does. Paste this into Claude Code, Codex, Grok, Gemini CLI or any agent that runs commands: "Read arena-api.greyforge.tech/age… and build me a Sley Arena bot. Test it until it beats the practice bots, then log me in and submit it." Free to play.
Sley Arena card with a FREE TO PLAY badge at the top right. Headline: Paste one prompt. Your agent builds the bot. A prompt box reads: Read https://arena-api.greyforge.tech/agent.md and build me a Sley Arena bot. Test it until it beats the practice bots, then log me in and submit it. Footer: Claude Code, Codex, Grok, Gemini CLI or any agent that runs commands. Bring a GitHub account. sleylang.org/arena

ALT Sley Arena card with a FREE TO PLAY badge at the top right. Headline: Paste one prompt. Your agent builds the bot. A prompt box reads: Read https://arena-api.greyforge.tech/agent.md and build me a Sley Arena bot. Test it until it beats the practice bots, then log me in and submit it. Footer: Claude Code, Codex, Grok, Gemini CLI or any agent that runs commands. Bring a GitHub account. sleylang.org/arena

1
3
1
5
188
Why should your AI write Python? 🤔 Python, JS, and Rust were made for people. Sley was made for AI: 🔒 Runs the same way, every time 🛑 Can never spin out of control 💸 Costs your agents no more to write than Python The language AI was waiting for. 🔥 👇 sleylang.org
Sley 2 - Machine-native programming for AI agents

Sley 2 - Machine-native programming for AI agents

Agents write typed Python. Sley checks every change against verified program state and commits only what passes. Sley 2.1.7 is self-hosted and open source under Apache-2.0.

sleylang.org
Sley (@SleyLanguage) profile pictureSley@SleyLanguageOct 21
Your AI agent's new favorite language just got cheaper. 💸 Sley 2.1.4: the agent write→test→fix loop costs ~3x fewer tokens than 2.1.3, and beats Python by 27% on typo fixes. Plus loops up to 2x faster. Same reproducible build. 👇 github.com/sley-lang/sley/re…
Release Sley 2.1.4 · sley-lang/sley

Release Sley 2.1.4 · sley-lang/sley

Sley 2.1.4 is a point release. The VM's compact scalar plan runs loops faster: the loop benchmarks are 1.56-1.95x faster than in 2.1.3, and call-bound fib(30) is unchanged. Sley is still about ...

github.com
1
1
1
4
225
Load more