🤖 11+ Models Available

All models,
one interface

Powered by NVIDIA NIM infrastructure. Switch models mid-conversation with a single click.

GLM · v5.1-0520 · 130B params

Best-in-class reasoning with native chain-of-thought. Excels at math, coding, analysis, and complex problem-solving. Flagship model for Axecodi.

reasoningcodemathmultilingual

Use GLM-5.1 →

DeepSeek · R1-671B · MoE

Advanced reasoning model with reinforcement learning. Matches o1-level performance on math and science tasks. Full chain-of-thought output.

reasoningmathscienceopen-source

Use DeepSeek R1 →

Meta · Llama-3.3-70B-Instruct

Meta's most powerful open-source LLM. Exceptional instruction following, coding, and multilingual support. Great for general tasks.

generalcodemultilingualopen-source

Use Llama 3.3 →

Mistral · Mixtral-8x7B-Instruct-v0.1

Mixture-of-Experts architecture with 8 expert networks. High throughput and efficiency. Excellent for rapid iteration and real-time applications.

fastMoEopen-source

Use Mixtral 8x7B →

Google · Gemma-2-9B-IT

Google's lightweight but capable open model. Punches above its weight class for summarization, Q&A, and chat. Ideal for lightweight tasks.

lightweightfastopen-source

Use Gemma 2 →

Alibaba · Qwen2.5-72B-Instruct

Alibaba's flagship open model with strong multilingual capabilities. Excellent for Chinese/English tasks, coding, and complex instruction following.

multilingualcodeopen-source

Use Qwen 2.5 →

Meta · CodeLlama-70B-Instruct

Specialized code generation model. Trained on 500B tokens of code. Supports 80+ programming languages with fill-in-the-middle capability.

codefill-in-middle80+ langs

Use CodeLlama 70B →

Mistral · Mistral-7B-Instruct-v0.3

Ultra-fast, lightweight model for quick queries and prototyping. Surprisingly capable at general tasks despite its compact size.

ultra-fastlightweightopen-source

Use Mistral 7B →

Model comparison

Model	Context Window	Speed	Reasoning	Code	Open Source
GLM-5.1	128K	Fast	✓ CoT	✓	✗
DeepSeek R1	64K	Moderate	✓ CoT	✓	✓
Llama 3.3 70B	128K	Fast	✓	✓	✓
Mixtral 8x7B	32K	Very Fast	Partial	✓	✓
Gemma 2 9B	8K	Ultra Fast	✗	Partial	✓
Qwen 2.5 72B	128K	Fast	✓	✓	✓