Model Releases
+1 to this. I was recently on a cross-continental flight without wifi, so I brought up Qwen3.6 & Gemma 4 (via @ollama) in Deep Agents on my …
+1 to this. I was recently on a cross-continental flight without wifi, so I brought up Qwen3.6 & Gemma 4 (via @ollama) in Deep Agents on my laptop. admittedly, they fell over on some more involved/com
+1 to this. I was recently on a cross-continental flight without wifi, so I brought up Qwen3.6 & Gemma 4 (via @ollama) in Deep Agents on my laptop. admittedly, they fell over on some more involved/complex prompts, but got a lot further along without requiring intervention than I had expected. If you haven't tried local models in the past month or two, (or gave up after the llama 3 days), it's worth revisiting! Deep Agents CLI discovers models on your machine automatically, so it's easy to swap with /model. If you feel like the model is reaching a ceiling, hot-swap mid-session to switch to a more capable/cloud model https://docs.langchain.com/oss/python/deepagents/cli/ Localmaxxing : pushing more inference to local models. Over five weeks, I tested how much of my daily work can run on a local 35B model instead of cloud frontier models. The answer : half. Many reasons to use local models : privacy, cost, asset depreciation. But the only one that…
Source: Harrison Chase (X) | 2026-05-11