Open-weights 35B-A3B agentic model post-trained from Qwen3.6-35B-A3B for search, coding and scientific research.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
BigBang-v1 is an experimental agentic language model built on Qwen3.6-35B-A3B, a mixture-of-experts model with 35B total and 3B active parameters. According to a YouTube review, it was post-trained on data produced by generator, critic and metacritic agents. It is released openly by endless-frontier; the vendor is not otherwise identified in the provided material.
The GitHub repository reports results on eight benchmarks covering long-horizon search (BrowseComp, xbench), software engineering (SWE-Bench Pro, SciCode-Verified), scientific research (FrontierScience-Research, HLE, BioMysteryBench) and AI research (MLE-Bench Lite, PaperBench Code-Dev). The authors state it scores highest among the compared 35B models and exceeds DeepSeek V4 Pro Preview on several of them. These figures are self-reported and use the authors' own harness and decontamination policy, so they may differ from official reports.
The repository provides the reported results, a lightweight general-agent harness with search, visit and code_exec tools, and download scripts for the SWE-Bench Pro and SciCode-Verified datasets. It is aimed at researchers and developers running agentic workflows locally or on their own infrastructure. The Hugging Face model card is mentioned in the README, but no URL was provided.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Experimental language model based on Qwen 3.6 35B, post-trained on data from generator, critic and metacritic agents. Scores much higher than the base on benchmarks such as Frontier Science; released openly.
Closest neighbours by meaning, not just by category.
Model
35B Mixture-of-Experts open-weights model from InternScience for long-horizon agentic search, engineering and tool use.
Model · Meituan
Meituan's open-weights 1.6T-parameter mixture-of-experts LLM (48B active), aimed at agentic coding and reasoning tasks.
Model · Alibaba
Alibaba's open-weights 2.4T-parameter MoE language model (95B active) from the Qwen3.8 family, with a hosted Max version.
Model
Open 7.39B dense model for math reasoning and agentic search, with a 256K context and the full training pipeline released.
Model · Tencent
Tencent's open-weights 295B Mixture-of-Experts model (21B active) for reasoning and agentic coding.
Model · Alpha Innovator
35B mixture-of-experts open-weights agentic model for engineering, long-horizon search and scientific research.