AI inference built for production.
SF
Joined August 2026
- Tweets13
- Following1
- Followers9
- Likes17
Pinned Tweet
Introducing Standard One SH: open-weight decision models.
One pass. A probability per option. No text generation.
Decision Index 0.3 public index:
- 3B 51.5, highest under 4B vs the Oct 12 board;
- 8B 57.4
Self-scored with the official kit.
Open weights on @huggingface
🧵 1/6
Jev-level decisions from an open model you can run yourself.
Two weeks ago we self-scored Standard One. This time the Decision Index board ran it: 35B SH 60.1, Jev 1.14 60.2, on private tests. ~3B active params per token, one GPU.
huggingface.co/StandardThink…
ALT Scatter plot of Decision Index 0.4.1 Full score against model size for 244 entries. Standard One 35B SH scores 60.1, on the Jev 1.14 reference line at 60.2.
More from the board run: above Jev 1.13 (59.2), #3 among MoE entries, best of the Qwen3.6-35B-A3B fine-tunes.
Full score = private tests only (60% same skills, 40% new domains); public benchmarks not counted.
Board: huggingface.co/spaces/multim…
Standard Thinking retweeted
good to see @standardthinkin being a leading 3B decision model in this comparison!
Standard Thinking retweeted
I've been spending my time in the past three months with @standardthinkin team and @jordanyechan their founder and CEO.
What they achieved with a small team & pre-seed resources is amazing - while hustling and iterating at the speed of light.
Standard One is live on HuggingFace and got over 24k downloads in the last 2-3 weeks.
Check out the Doom demo - I will put it in the comment 👇 As they say, it's not a computer if it cant run Doom.
Let me know if you need an intro - we are going to be pushing more fun stuff in the coming weeks.
Introducing Standard One SH: open-weight decision models.
One pass. A probability per option. No text generation.
Decision Index 0.3 public index:
- 3B 51.5, highest under 4B vs the Oct 12 board;
- 8B 57.4
Self-scored with the official kit.
Open weights on @huggingface
🧵 1/6
Standard One 3B SH v2 now ranks #1 among sub-4B models on the Jev Decision Index by @multimodalart (Full score: 39.9).
3B v3 is already out. We're excited to see where it lands on the leaderboard.
ALT Decision Index 0.3.2 leaderboard, TEXT, Full score, models under 4B parameters, Jev hidden: Standard One 3B SH (v2) at 39.9, followed by 38.9, 38.9 and 38.6.
Introducing Standard One SH: open-weight decision models.
One pass. A probability per option. No text generation.
Decision Index 0.3 public index:
- 3B 51.5, highest under 4B vs the Oct 12 board;
- 8B 57.4
Self-scored with the official kit.
Open weights on @huggingface
🧵 1/6
Sources:
Leaderboard (Decision Index 0.3.2 · TEXT · Full score · ≤4B): huggingface.co/spaces/multim…
3B SH weights, v3, on @huggingface: huggingface.co/StandardThink…
One photo of each parcel, two options: pass, or review for damage. Of 30 parcels One 8B passed 28 and sent 2 to review.
The answers are real; the conveyor footage is AI-generated, shown at 2.5× speed. Recorded earlier on One 8B, the API model of the time.
5/6
Open weights, Apache-2.0:
8B SH: huggingface.co/StandardThink…
3B SH: huggingface.co/StandardThink…
Try SH in the API, free credit: standardthinking.ai/standard…
Benchmarks: standardthinking.ai/resource…
Self-scored with the official kit; submission under review.
6/6
