🐼🌿 We're a bamboo-munching, crypto-loving team of pandas who've traded the forest for the Ethereum blockchain! We were born during #TestingTheMerge
- Tweets 229
- Following 8
- Followers 2,512
- Likes 126
Introducing the Platåberget testnet! Its our Glamsterdam focused testnet to help the community test all the upcoming changes. Please get involved with the testing efforts!
blog.ethereum.org/2026/08/17…
5
38
116
22,829
The Platåberget testnet is one of the last steps before we start deploying changes on long lived testnets and eventually Mainnet. So Glamsterdam is close!
1
12
204
The new Deepseek V4 flash 0731 ranked highly on our ethiq leaderboard greatly improving on execution specs +17 and consensus-specs +15 questions over the previous checkpoint.
We self hosted the model with thinking set to Max.
It even beat out Claude Opus 5 due to refusals.
1
4
29
1,923
We built Panda, a CLI/MCP tool that lets agents work directly with our observability data.
We found that if you give the agent a code sandbox and a library for your stack, things start to click together very nicely.
Post below 👇
2
4
25
2,526
We've learned some lessons along the way. This one is a must read for anyone seriously using agents (even outside of Ethereum!)
ethpandaops.io/posts/panda/
5
578
Claude Opus 4.8 takes the top spot on EthIQ with a score of 88%!
Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.
Available today at the same price.
ALT Benchmark table showing how Claude Opus 4.8 compares to its predecessor and to other models on tests of coding, agentic skills, reasoning, and practical knowledge work tasks.
1
3
34
3,113
It turns out some AI models can step through EVM bytecode in their head.
We built EthIQ, a new benchmark to test how well models actually understand Ethereum protocol internals.
325 questions across two evaluation modes, and one alive canary 🐦
Read more 👇
31
32
215
20,801
We tested 44+ models across EVM execution, consensus state transitions, fork choice, and more. Anti-memorization is baked in. Inputs are mutated from the official spec fixtures so training data is ruled out.
The results were... surprising.
Check it out: ethiq.ethpandaops.io
1
1
24
1,551
Read the blog post for all the details here: ethpandaops.io/posts/introdu…
1
1
25
1,259