Qwen3.6 35B A3B abliterated v4

Qwen3.6

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released June 2026262K context≈ 196,608 words

A community-modified variant of Qwen's 35B mixture-of-experts model, with abliteration applied to remove refusal behaviors. It activates roughly 3B parameters per forward pass despite its larger total parameter count, keeping inference costs relatively low. The abliteration process means it responds without built-in content restrictions, which shifts responsibility for safety guardrails entirely to the developer.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Long Context

Exceptional

Coding

Qwen3.6 35B A3B abliterated v4

Qwen3.6

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released June 2026262K context≈ 196,608 words

A community-modified variant of Qwen's 35B mixture-of-experts model, with abliteration applied to remove refusal behaviors. It activates roughly 3B parameters per forward pass despite its larger total parameter count, keeping inference costs relatively low. The abliteration process means it responds without built-in content restrictions, which shifts responsibility for safety guardrails entirely to the developer.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Long Context

Exceptional

Coding

Glossary

AbliterationA technique that removes or disables a model's built-in safety refusal mechanisms, allowing it to respond to a wider range of requests.Content RestrictionsSafety guidelines and filters built into a model to prevent it from generating harmful, illegal, or unethical content.Forward PassA single computation cycle where input data flows through the model's layers to produce an output prediction.GuardrailsSafety mechanisms built into a model to refuse harmful requests or prevent it from generating unsafe content.InferenceThe process of running a trained model to generate predictions or outputs from new inputs.Parameter CountThe total number of adjustable weights in a model; more parameters generally mean more capacity to learn, but also require more computing power.ParametersThe learned numerical values in a model — more parameters generally means more capacity but higher compute cost.Safety GuardrailsBuilt-in restrictions or filters that prevent a model from generating harmful, illegal, or unethical content.

Capabilities

Capabilities

Use Case Fit

Glossary