by IQuest Lab
320B-parameter MoE model (15B active) from IQuest for agentic coding and tool use, with 512K context.
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.
IQuest-Q1 is a Mixture-of-Experts language model from IQuest Lab built for agentic coding, reasoning and multi-step tool use. It has about 320B total parameters with an estimated 15B activated per token, 88 transformer layers, 256 experts (8 active), and a hybrid attention pattern of three sliding-window layers to one full-attention layer. The context length is 524,288 tokens. It also includes multi-token prediction (MTP) layers used for speculative decoding at inference.
The model is intended to be driven by CLI coding agents such as Claude Code and Codex. The README recommends Claude Code 2.1.140 or Codex 0.142, with sampling at temperature 1.0, top-p 0.95 and top-k 20. Tool calling goes through a gateway that supports the Anthropic Messages API (Claude Code) or the OpenAI Responses API (Codex). The authors report benchmark results against DeepSeek-V4 models, including an in-house CLI benchmark, but the figures are self-reported and not independently verified here.
It is self-hosted: the repository provides Docker images and launch commands for SGLang and a vLLM fork, exposing an OpenAI-compatible API, with the examples using 8-way tensor parallelism. The model is text-only; it has no multimodal input. The repository license is listed as NOASSERTION, so the exact terms should be checked before use. The Hugging Face repository is referenced in the README's download commands (IQuestLab/IQuest-Q1).
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
An agentic MoE model for CLI tools like Claude Code and Codex, with 320B total and 15B active parameters. It works through long software tasks and performs well for its size. Code is released.
Closest neighbours by meaning, not just by category.
Model · Z.ai
Z.ai's open-weights 753B MoE language model, a post-trained GLM-5.2 aimed at coding, agentic tasks and cybersecurity.
Model · Meituan
Meituan's open-weights 1.6T-parameter multimodal MoE model (48B active) for agentic coding and visual 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
35B Mixture-of-Experts open-weights model from InternScience for long-horizon agentic search, engineering and tool use.
Model · Accio Lab
Open-weights 35B MoE (3B active) agentic model from Accio Lab, post-trained from Qwen3.6-35B-A3B.
Model · Meituan
Meituan's open-weights 1.6T-parameter mixture-of-experts LLM (48B active), aimed at agentic coding and reasoning tasks.