llm-selector

Choose the right LLM, then verify you’re actually using it.

Free, practical guidance for SMEs, OPCs, freelancers, and small teams choosing between GLM, Qwen, DeepSeek, Kimi, Claude, GPT, and local open-source models.

This blog publishes our free solutions and case studies. Everything here is free to read and reuse.

Start here

• Free consultation case studies — real public questions, turned into practical recommendations
• Daily hot topics + free solutions — recurring digest of LLM selection / access / cost / deployment problems with free fixes
• Decision methodology — how we think about LLM selection
• Vision and go-to-market — where this is going
• Market validation — public demand signals and investor-readable cases

What this is

Most LLM content answers “which model scores higher on a benchmark?”

We answer the more practical questions:

• Which model should we use for AI coding, documents, customer service, or knowledge base?
• Hosted API, domestic provider, hybrid routing, or private deployment?
• What risks should we review before sending company data to a model?
• If a gateway claims to route to GLM-5.2, how do we verify it?
• What should our first proof-of-concept look like?

Try it

git clone https://github.com/xiaoyanfei-tech/llm-selector.git
cd llm-selector
python selector.py

Get a free first-pass selection

Open a selection request: github.com/xiaoyanfei-tech/llm-selector/issues/new/choose

Include your scenario, data sensitivity, budget range, deployment preference, and current tools. Do not include API keys, private URLs, customer records, contracts, or confidential screenshots.

Related project

verify-glm — verify whether the model behind your Claude Code / LLM gateway is actually the model you think it is.