Qwen (Alibaba) models

Alibaba's newest. Frontier coding, math, and reasoning with built-in thinking; a hybrid attention design keeps long documents fast. The strongest model here for 24GB-class graphics cards.

Top-tier coding, math, and reasoning. The 35B MoE runs fast for its size — only 3B parameters active per token.

Alibaba's agentic coding model — a 30B MoE with only 3B active per token. Tuned for repository-scale work and tool use (Qwen Code, Cline).

R1 reasoning distilled onto a Qwen3 8B base by DeepSeek themselves

Powerful chain-of-thought reasoning. Excels at complex problem solving, math, and logic.

Built on Qwen 3.5

DeepReinforce's agentic coding model, self-improved with RL - 1.5 extends the loop to generating its own training tasks. State-of-the-art among open coders at its size, purpose-built for terminal coding agents and tool use. The 9B runs on modest machines; the 35B is a mixture-of-experts (3B active per token) that runs fast for its size and adds vision.

Qwen 3.5 distilled with Claude Opus reasoning traces

Qwen 3.5 distilled from Claude Opus reasoning traces. Enhanced chain-of-thought capabilities.

Qwen 3.5 with safety training removed

Qwen 3.5 with safety guardrails removed. No content filtering or refusals.

Distilled from Qwen 3.8

Empero's full-parameter distillations of the Qwen 3.8 flagship into small, fast sizes - a big knowledge jump over same-size models (the 9B scores near the giants on broad knowledge), with step-by-step reasoning. Text only.

Qwen 3.8 with safety training removed

The Qwen 3.8 flagship with refusal behavior substantially reduced (openly documented - reduced, not eliminated). Same frontier coding, math, and reasoning, and it can see images with its vision add-on. For 24GB-class graphics cards.

A Qwen 3.5-based merge

A community Qwen 3.5-based merge — uncensored and multimodal (it can see images you attach), with step-by-step reasoning and tool use. A creative, unfiltered generalist. v2 trains out the repetition loops of the original.