QuanLLM-v1.0-qm is a major upgrade of the QuanLLM field branch for quantum mechanics teaching and application. Now built on Qwen3-8B, it adds a visible thinking mode, streaming output, and SymPy symbolic computation: the model plans the derivation, SymPy guarantees the math. Ships with a no-install CLI client for Windows, macOS, and Linux.
How to apply: send an email to activity@quanllm.com following the fixed template below.
v0.2-qm moved the model to the cloud. v1.0-qm fixes the two remaining weaknesses of a language model in physics teaching: the accuracy of symbolic math and the transparency of reasoning. Computation is done by the SymPy engine, exact down to the expression itself; thinking mode lets the model reason first and answer second, so students see how a problem is approached and teachers can check where it goes right or wrong.
Targets high-frequency final exam topics: concept explanation, formula derivation, calculation problems, and proof questions.
Not limited to exam questions; also explains concepts, compares representations, and discusses physical pictures and common misconceptions.
Reason first, answer second, with the full chain of thought visible. Both thinking and answers stream token by token—no more waiting on a blank screen.
Differentiation, integration, and equation solving are delegated to the symbolic engine, exact to the expression, with multi-round tool calls in a single conversation.
Explains core concepts such as wave functions, the uncertainty principle, representation transformations, identical particles, spin, and entanglement in clear language.
Step-by-step derivations of the Schrödinger equation, ladder operators, angular momentum commutation relations, perturbation theory, and more.
Problem-solving strategies and step-by-step solutions for harmonic oscillators, hydrogen atoms, perturbation theory, and scattering—with SymPy guaranteeing every computation.
Answers open-ended quantum mechanics questions, including physical pictures, comparisons of theoretical frameworks, and connections to frontiers.
No-install client for Windows, macOS, and Linux, with multi-turn conversation and session history save & restore. Ready out of the box.
Every v1.0-qm API Key ships with a self-adapting command-line client. No key is embedded in the program—an APIKEY template file is generated on first run. Streaming, visible thinking, and SymPy tool calls are all preserved. Only 50 API Keys are available in the first batch.
To download the release for your platform (Windows / macOS / Linux), just click the GitHub button in the top-right corner.
Only 50 spots are available in the first batch. Please send an email using the fixed template below. We will reply with the API Key, the three-platform client packages, and usage instructions after review.
⚠️ Please make sure all fields above are filled out completely. Missing key information may delay the review process. Please do not send duplicate applications from the same email address or applicant.
The current demo is static. Full conversational functionality requires an API Key, after which you can chat through the three-platform client.
💡 Tip: Apply for an API Key to start conversing with the model through the three-platform client. Please follow the application template and send an email to activity@quanllm.com. This page is for product demonstration only.
v1.0-qm is opening 50 API Keys in its first batch. Approved applicants receive the three-platform client, ready out of the box. Feedback from students, teachers, and researchers is welcome.