Top 100 AI · United Kingdom
Peter Gostev
Independent 
- Score
- 6,516−16 today
- Rank
- 104/198 ↓16
- over 90 days
- −21−0.3%
Peter Gostev is an AI Capability Lead at Arena.ai, focusing on generative AI and benchmarking models. He publishes detailed evaluations of AI models, particularly focusing on their ability to handle nonsensical queries through his 'Bullshit Benchmark'. His content includes technical demonstrations, comparisons of AI models, and insights into AI development environments.
Data as of Oct 25, 2026.
The raw numbers · last 90 days
- Median views
- 14.8K
- per post
- Median engagements
- 185
- per post
- Posts
- 398
- over 90 days
- Followers
- 97.1K
- all scored platforms
What drives the score
Percentiles 0–100 against all tracked creators.
- Reach · 30%73
Average views per post over 90 days
- Engagement · 30%55
Average reactions per post
- Audience · 30%70
Total followers, relative to population
- Consistency · 10%100
Posting frequency (25 posts / 90 days = 100)
Platforms
97.1K total followers
| Platform | Followers | Score |
|---|---|---|
| X@petergostev | 29K | 7,034t3 |
| LinkedIn@peter-gostev | 68.1K | 9,628t1 |
Last 30 days
Gostev's top posts
- X · 20 days ago#1
Mistral Large 4 was trained on 4,000 GPUs and Astra was trained on 100,000 of them. We'll see what the kitten can pull off, but the disparity is pretty stark
PG168K views3K engagements
X · 12 days ago#2PGHoly cow, so based on OpenAI's UK filings, average salary in 2025 was £794k and £1.52m share based compensation on top of that. So the total was £2.35m = $3.1m per employee https://t.co/YOUYMmTBkg
615.5K views2K engagements
X · 23 days ago#3PGI have a Continuous Learning benchmark where models attempt to learn to play chess. They are given a /goal of learning and improving playing against a Stockfish opponent in 200 games. They can choose the difficulty, t…
2.2M views2K engagements
- X · 20 days ago#4
"On average, each result used three hours of ChatGPT Pro thinking compute with [the unreleased internal OpenAI] model." So these unsolved problems were solved in 3 hours each of next gen chatty pro? 💀
PG84.9K views1.9K engagements
- X · 30 days ago#5
Calling LLMs "stochastic parrots" in 2021, then claiming Fable and Astra aren’t LLMs, is like arguing in 1900 that planes would never fly, then when they carry millions people, arguing that: "they aren’t really planes…
PG68.2K views1.7K engagements
X · 20 days ago#6PGI probably spent 300 billion tokens on Hadwiger–Nelson problem (GPT-5.6 and Astra) and got absolutely nothing - and here it is, five-colours is no go, lower bound has moved to 6 or 7. Not a full solution yet, but it's…
99.5K views1.4K engagements
Head-to-head
