ERNIE 5.1

ERNIE 5.1 Thinking

ERNIE 5.1 Thinking is the latest Wenxin model with explicit reasoning enabled for harder planning, coding, math, and agentic tasks. ⚠️ Note: privacy and logging guarantees may be limited.

Ernie 5.0 Thinking Preview

ERNIE 5.0 Thinking is Baidu's next-generation step-by-step reasoning model, providing stronger logical thinking, code generation, and tool-use capabilities with explicit thinking traces. ⚠️ Note: This model routes through Baidu, a Chinese entity - privacy and logging guarantees may be limited.

Gemma 4 Dark Thoughts V2 31B

Ateron's second Dark Thoughts merge combines Dark Scarlett V2, MeroMero V2, and Scotoma 2 for creative dialogue and extended character roleplay. This text-only Gemma 4 31B deployment supports a 65,536-token context window and up to 8,192 output tokens.

DeepSeek V4 Flash Vision Exp Cybersecurity

A cybersecurity-focused variant of DeepSeek V4 Flash Vision Exp, with image understanding, tool calling, optional reasoning, and a 1,048,576-token context window.

Gemini 4 Argon

Google's Gemini 4 Argon is a fast multimodal model for agentic workloads, including coding, tool use, image understanding, PDF and document extraction, audio, and video.

Gemma 4 Gutenberg 31B

nbeerbower's Gemma 4 31B finetune for literary fiction, character interiority, and controlled narrative pacing. Its text decoder was trained with ORPO on 14,816 Athanorlite-DPO preference pairs. This text-only deployment supports a 65,536-token context window and up to 8,192 output tokens.