Do Qwen3.6 27B quantizations break the pelican?
We tested Qwen3.6 27B quantizations by Unsloth on Hugging Face, with pelicans on bikes, gears, Terminal-Bench 2.1, and AIME-120.
Piotr Migdał
Read more
Optimize your AI costs
Full quality, fewer tokens.
We find what burns your team’s AI budget and fix it.
Let’s talkBook a direct call with the CEOObserve
Where every token goes: by model, team, and use case.
Organize
Budgets and limits per team. No end-of-month surprises.
Optimize
Right-size models, tighten prompts, cut context. Evals keep quality in check.
Latest insights
Explore our research on AI agents, benchmarking, and evaluation
I burned all my tokens researching how to save tokens
How I burned a full Claude limit in 30 minutes and built my own deep research pipeline instead: 3 subscriptions, shared memory, a clear role for each model. You can build the same from what you already pay for.
Bartosz Kotrys
Read more
Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost?
We ported the puzzle game Baba Is You to the Harbor framework, and benchmarked current models, including Claude, GPT, Gemini, GLM, and DeepSeek. A human Twitch streamer is 4x faster than Claude Fable 5.
Piotr Migdał & Piotr Grabowski
Read more