Qwen 3.8 Flash just hit Atomic Agent.
A FREE local agent - $0 per token, runs on your machine. No cloud, no API key. Inference is local GBNF grammar constraints force valid tool calls from any model KV-cache reuses compute across steps. SQLite stores searchable session
Qwen3.8-Flash-Next from Alibaba_Qwen has day-0 support in Atomic Agent!
125B main model, 51B N-gram embeddings, 6B activated per token and 62.5 on SWE-bench Pro. We gave it a folder and it checked every file to sort them by content. Total: 25 steps + 230K tokens ≈ $0.07
DeepSeek just turned V4 Flash into a vision-powered agent.
Not “upload an image and get a description.” It can see a screenshot, reason about what’s happening, use tools, and act on it.
GLM-5.3-Flash vs. DeepSeek-v4-Flash
Same prompt, same harness (pi), same effort. GLM used 114k tokens over 93 turns taking 73min. Total cost 16 cents. DeepSeek used 70k tokens over 9 turns taking 10min. Total cost 2 cents. GLM is a richer scene but at 8x the cost and 7x the
You have reached the end of the archive
All of deepseek