Models

A million tokens of context, tool calling and vision, with no refusals. Point a coding agent at it.

$0.30 in · $1.00 out · 1M context

Kimi K3.1 Pro, uncensored: the strongest model on the line, with a million tokens of context, tool calling, reasoning and vision.

$7.50 in · $37.50 out · 1M context

The full GLM 5.3, uncensored: frontier reasoning, a million tokens of context and tool calling, with no refusals. Text only: it does not take images.

$2.00 in · $5.00 out · 1M context

DeepSeek V4.1 Flash 1017, the October update, uncensored: a million tokens of context, tool calling, reasoning and vision, with no refusals.

$1.25 in · $2.50 out · 1M context

Xiaomi MiMo with the refusal direction removed. A million tokens of context, tool calling and vision.

$1.00 in · $2.00 out · 1M context

A small, cheap uncensored Qwen with a million tokens of context and tool calling. The budget pick for agents.

$1.50 in · $5.00 out · 1M context

GLM 5.3 Flash tuned for security work: code review, exploit analysis and incident response. A million tokens of context, tool calling and vision. Security-tuned, not an uncensored model.

1M context