Ollama ExplorerBeta

Codeintermediate

dolphin-mixtral

Dolphin

Uncensored, 8x7b and 8x22b fine-tuned models based on the Mixtral mixture of experts models that excels at coding tasks. Created by Eric Hartford.

1.1M pullsUpdated Feb 26, 202570 tags64K context

Quick start

ollama run dolphin-mixtral

Available sizes

Tag	Size	Quantization	Context	Min RAM
dolphin-mixtral:latest	26GB	q4_k_m	32K context	32.5 GB
dolphin-mixtral:8x22b	80GB	q4_k_m	64K context	100 GB

Strengths & Limitations

Strengths

Coding tasks
Mixture of experts
Uncensored

Related models

qwen2.5-coderCode

The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing.

A large language model that can use text prompts to generate and discuss code.

qwen3-coderCode

Alibaba's performant long context models for agentic and coding tasks.

deepseek-coderCode

DeepSeek Coder is a capable coding model trained on two trillion code and natural language tokens.