Decide Live
๐ฏ
38
On-device decisions from camera and microphone
Proto_AGI
mayafree
AI & ML interests
None yet
Recent Activity
liked a Space 2 days ago
FINAL-Bench/Decision-AI-Web liked a Space 2 days ago
masterset-ai/PhysicalEye-Decide-35B-demo liked a dataset 2 days ago
FINAL-Bench/oqc-dataOrganizations
liked 2
Spaces 2 days ago
liked a
dataset 2 days ago
upvoted an article 2 days ago
Article
The Open Quantum Challenge: Quantum Simulation and QEC Decoding on Classical GPUs
FINAL-Bench
โข โข 11 liked a
Space 2 days ago
reacted to SeaWolf-AI's post with ๐ค 2 days ago
Post
4546
โ๏ธ The Open Quantum Challenge is live. Both seasons open today.
Quantum computers are expensive, queued, and reachable by only a few. So we removed the barrier. With a single GPU and a public harness, anyone can contribute to core quantum-computing problems with no quantum hardware at all. That is the point of this challenge: to move quantum research from a handful of labs to everyone.
โข Season 1, Quantum simulation: reproduce a target quantum system on a GPU.
โข Season 2, QEC decoder: correct errors on coded quantum states.
Answers are held privately and every submission is auto-scored against a frozen ground truth, so the ranking is reproducible and hardware-independent.
๐ Prize: 2,000 USD total (1,000 USD per season, ๐ฅ600 / ๐ฅ300 / ๐ฅ100)
Getting started takes one step: copy the participation guide from the Space and paste it into Codex or Claude Code, then submit against the harness.
๐ Read the article: https://huggingface.co/blog/FINAL-Bench/the-open-quantum-challenge-quantum-simulation-and
๐ Challenge: FINAL-Bench/open-quantum-challenge
Quantum computers are expensive, queued, and reachable by only a few. So we removed the barrier. With a single GPU and a public harness, anyone can contribute to core quantum-computing problems with no quantum hardware at all. That is the point of this challenge: to move quantum research from a handful of labs to everyone.
โข Season 1, Quantum simulation: reproduce a target quantum system on a GPU.
โข Season 2, QEC decoder: correct errors on coded quantum states.
Answers are held privately and every submission is auto-scored against a frozen ground truth, so the ranking is reproducible and hardware-independent.
๐ Prize: 2,000 USD total (1,000 USD per season, ๐ฅ600 / ๐ฅ300 / ๐ฅ100)
Getting started takes one step: copy the participation guide from the Space and paste it into Codex or Claude Code, then submit against the harness.
๐ Read the article: https://huggingface.co/blog/FINAL-Bench/the-open-quantum-challenge-quantum-simulation-and
๐ Challenge: FINAL-Bench/open-quantum-challenge
reacted to SeaWolf-AI's post with ๐๐๐ 3 days ago
Post
4079
๐ง We just released Darwin-27B-ZTC, a judgment engine that reaches a verdict without generating anything.
Most LLMs answer by generating, decoding one token at a time. Darwin-27B-ZTC takes a different route.
โ๏ธ How it works
๐น It makes its call in a single forward pass.
๐น Zero generated tokens, and no decoding loop.
๐น That keeps latency and cost far below what a generative model needs.
๐ฏ What it judges
๐น It handles several question types: free-form correctness (noul), multiple choice (choice), and scoring (score).
๐น For each one it hands back a calibrated confidence, not just an answer.
๐ How well calibrated (measured)
๐น KL 0.204, Brier 0.097, so the confidence it reports lines up with what actually happens.
๐น 0.743 accuracy (zero-shot, general split), across 2,000 judgments with zero errors.
๐น By type: noul 0.847, choice 0.723, score 0.675.
๐น None of the benchmark's train split went into it. It is pure zero-shot.
๐ Where it fits
๐น Grading at scale, model routing, safety gating, anywhere you want a fast decision without paying for generation.
๐ It currently sits at #1 on the official typed-decisions leaderboard on Hugging Face (0.743 accuracy, zero-shot).
๐ Links
Model: FINAL-Bench/Darwin-27B-ZTC
Leaderboard: LocalLLaMA/typed-decisions
Curious to hear what you make of the single-pass, no-generation approach. ๐
Most LLMs answer by generating, decoding one token at a time. Darwin-27B-ZTC takes a different route.
โ๏ธ How it works
๐น It makes its call in a single forward pass.
๐น Zero generated tokens, and no decoding loop.
๐น That keeps latency and cost far below what a generative model needs.
๐ฏ What it judges
๐น It handles several question types: free-form correctness (noul), multiple choice (choice), and scoring (score).
๐น For each one it hands back a calibrated confidence, not just an answer.
๐ How well calibrated (measured)
๐น KL 0.204, Brier 0.097, so the confidence it reports lines up with what actually happens.
๐น 0.743 accuracy (zero-shot, general split), across 2,000 judgments with zero errors.
๐น By type: noul 0.847, choice 0.723, score 0.675.
๐น None of the benchmark's train split went into it. It is pure zero-shot.
๐ Where it fits
๐น Grading at scale, model routing, safety gating, anywhere you want a fast decision without paying for generation.
๐ It currently sits at #1 on the official typed-decisions leaderboard on Hugging Face (0.743 accuracy, zero-shot).
๐ Links
Model: FINAL-Bench/Darwin-27B-ZTC
Leaderboard: LocalLLaMA/typed-decisions
Curious to hear what you make of the single-pass, no-generation approach. ๐
- 2 replies
liked a
Space 4 days ago
reacted to SeaWolf-AI's post with ๐ 4 days ago
Post
4102
๐ Darwin-27B-ZTC-v2 just took #1 on the System One Mosaic Benchmark (S1MB).
S1MB compares 102 models across 137 specialized benchmarks, in three task types: Noul (assess a condition), Choice (select an option), Score (rate on a scale). Ranking is by overall Borda score.
๐ Top of the board
๐ฅ Darwin ZTC v2 (FINAL-Bench) 89.58
๐ฅ OpenJev-27B 87.50
๐ฅ AutoJev-27B 87.07
4๏ธโฃ Eikos 27B 85.43
5๏ธโฃ Jev 1.13 85.05
๐ Ranks 2 to 5 are all the JEV family (TypeSafe AI's System One model, from ex-OpenAI researchers). S1MB exists to compare these System One judges, so leading it is the headline.
โ๏ธ Why a zero-token judge wins here
๐น It does not generate. It reads the input and typed questions and returns a calibrated distribution in a single forward pass.
๐น Zero generated tokens, no decoding loop, so latency and cost stay low.
๐น Holds up out of distribution too: General Noul 96.00, General Choice 99.34.
It is also #1 on the typed-decisions leaderboard (0.743, zero-shot). Same message from both: a deterministic, calibrated judge at one forward pass per call.
๐ Model: FINAL-Bench/Darwin-27B-ZTC-v2
๐ Leaderboard: hotchpotch/S1MB-leaderboard
Standings move as new models are added. Numbers reflect the board at the time of writing. ๐
S1MB compares 102 models across 137 specialized benchmarks, in three task types: Noul (assess a condition), Choice (select an option), Score (rate on a scale). Ranking is by overall Borda score.
๐ Top of the board
๐ฅ Darwin ZTC v2 (FINAL-Bench) 89.58
๐ฅ OpenJev-27B 87.50
๐ฅ AutoJev-27B 87.07
4๏ธโฃ Eikos 27B 85.43
5๏ธโฃ Jev 1.13 85.05
๐ Ranks 2 to 5 are all the JEV family (TypeSafe AI's System One model, from ex-OpenAI researchers). S1MB exists to compare these System One judges, so leading it is the headline.
โ๏ธ Why a zero-token judge wins here
๐น It does not generate. It reads the input and typed questions and returns a calibrated distribution in a single forward pass.
๐น Zero generated tokens, no decoding loop, so latency and cost stay low.
๐น Holds up out of distribution too: General Noul 96.00, General Choice 99.34.
It is also #1 on the typed-decisions leaderboard (0.743, zero-shot). Same message from both: a deterministic, calibrated judge at one forward pass per call.
๐ Model: FINAL-Bench/Darwin-27B-ZTC-v2
๐ Leaderboard: hotchpotch/S1MB-leaderboard
Standings move as new models are added. Numbers reflect the board at the time of writing. ๐
reacted to SeaWolf-AI's post with ๐ฅ๐ 4 days ago
Post
4079
๐ง We just released Darwin-27B-ZTC, a judgment engine that reaches a verdict without generating anything.
Most LLMs answer by generating, decoding one token at a time. Darwin-27B-ZTC takes a different route.
โ๏ธ How it works
๐น It makes its call in a single forward pass.
๐น Zero generated tokens, and no decoding loop.
๐น That keeps latency and cost far below what a generative model needs.
๐ฏ What it judges
๐น It handles several question types: free-form correctness (noul), multiple choice (choice), and scoring (score).
๐น For each one it hands back a calibrated confidence, not just an answer.
๐ How well calibrated (measured)
๐น KL 0.204, Brier 0.097, so the confidence it reports lines up with what actually happens.
๐น 0.743 accuracy (zero-shot, general split), across 2,000 judgments with zero errors.
๐น By type: noul 0.847, choice 0.723, score 0.675.
๐น None of the benchmark's train split went into it. It is pure zero-shot.
๐ Where it fits
๐น Grading at scale, model routing, safety gating, anywhere you want a fast decision without paying for generation.
๐ It currently sits at #1 on the official typed-decisions leaderboard on Hugging Face (0.743 accuracy, zero-shot).
๐ Links
Model: FINAL-Bench/Darwin-27B-ZTC
Leaderboard: LocalLLaMA/typed-decisions
Curious to hear what you make of the single-pass, no-generation approach. ๐
Most LLMs answer by generating, decoding one token at a time. Darwin-27B-ZTC takes a different route.
โ๏ธ How it works
๐น It makes its call in a single forward pass.
๐น Zero generated tokens, and no decoding loop.
๐น That keeps latency and cost far below what a generative model needs.
๐ฏ What it judges
๐น It handles several question types: free-form correctness (noul), multiple choice (choice), and scoring (score).
๐น For each one it hands back a calibrated confidence, not just an answer.
๐ How well calibrated (measured)
๐น KL 0.204, Brier 0.097, so the confidence it reports lines up with what actually happens.
๐น 0.743 accuracy (zero-shot, general split), across 2,000 judgments with zero errors.
๐น By type: noul 0.847, choice 0.723, score 0.675.
๐น None of the benchmark's train split went into it. It is pure zero-shot.
๐ Where it fits
๐น Grading at scale, model routing, safety gating, anywhere you want a fast decision without paying for generation.
๐ It currently sits at #1 on the official typed-decisions leaderboard on Hugging Face (0.743 accuracy, zero-shot).
๐ Links
Model: FINAL-Bench/Darwin-27B-ZTC
Leaderboard: LocalLLaMA/typed-decisions
Curious to hear what you make of the single-pass, no-generation approach. ๐
- 2 replies
upvoted an article 4 days ago
Article
Leading the System One Mosaic Benchmark: What Darwin-27B-ZTC-v2's #1 Means
FINAL-Bench
โข โข 6 reacted to SeaWolf-AI's post with ๐ค๐๐ฅ 4 days ago
Post
4102
๐ Darwin-27B-ZTC-v2 just took #1 on the System One Mosaic Benchmark (S1MB).
S1MB compares 102 models across 137 specialized benchmarks, in three task types: Noul (assess a condition), Choice (select an option), Score (rate on a scale). Ranking is by overall Borda score.
๐ Top of the board
๐ฅ Darwin ZTC v2 (FINAL-Bench) 89.58
๐ฅ OpenJev-27B 87.50
๐ฅ AutoJev-27B 87.07
4๏ธโฃ Eikos 27B 85.43
5๏ธโฃ Jev 1.13 85.05
๐ Ranks 2 to 5 are all the JEV family (TypeSafe AI's System One model, from ex-OpenAI researchers). S1MB exists to compare these System One judges, so leading it is the headline.
โ๏ธ Why a zero-token judge wins here
๐น It does not generate. It reads the input and typed questions and returns a calibrated distribution in a single forward pass.
๐น Zero generated tokens, no decoding loop, so latency and cost stay low.
๐น Holds up out of distribution too: General Noul 96.00, General Choice 99.34.
It is also #1 on the typed-decisions leaderboard (0.743, zero-shot). Same message from both: a deterministic, calibrated judge at one forward pass per call.
๐ Model: FINAL-Bench/Darwin-27B-ZTC-v2
๐ Leaderboard: hotchpotch/S1MB-leaderboard
Standings move as new models are added. Numbers reflect the board at the time of writing. ๐
S1MB compares 102 models across 137 specialized benchmarks, in three task types: Noul (assess a condition), Choice (select an option), Score (rate on a scale). Ranking is by overall Borda score.
๐ Top of the board
๐ฅ Darwin ZTC v2 (FINAL-Bench) 89.58
๐ฅ OpenJev-27B 87.50
๐ฅ AutoJev-27B 87.07
4๏ธโฃ Eikos 27B 85.43
5๏ธโฃ Jev 1.13 85.05
๐ Ranks 2 to 5 are all the JEV family (TypeSafe AI's System One model, from ex-OpenAI researchers). S1MB exists to compare these System One judges, so leading it is the headline.
โ๏ธ Why a zero-token judge wins here
๐น It does not generate. It reads the input and typed questions and returns a calibrated distribution in a single forward pass.
๐น Zero generated tokens, no decoding loop, so latency and cost stay low.
๐น Holds up out of distribution too: General Noul 96.00, General Choice 99.34.
It is also #1 on the typed-decisions leaderboard (0.743, zero-shot). Same message from both: a deterministic, calibrated judge at one forward pass per call.
๐ Model: FINAL-Bench/Darwin-27B-ZTC-v2
๐ Leaderboard: hotchpotch/S1MB-leaderboard
Standings move as new models are added. Numbers reflect the board at the time of writing. ๐
liked a
model 4 days ago
liked a
Space 4 days ago
liked a
dataset 5 days ago