A common starting point
Students use a course-issued API key. No individual provider account or personal paid subscription is needed for course API access.
Department of Computer Science · Pilot proposal
An AI lab for
every CS student.
One shared API for the classroom, the lab, and the next big idea. Give West Chester students access to capable AI models—with course budgets and faculty in control.
Checking gatewayOpenAI-compatible APIInstructor-issued access
The shared AI classroom01 / Access architecture
WCUPA · CS AI Gateway
One endpoint. Many possibilities.
POST /v1/chat/completions
{
"model": "glm-4.7-flash",
"messages": [{ "role": "user",
"content": "Help me test my algorithm." }]
}
{
"model": "glm-4.7-flash",
"messages": [{ "role": "user",
"content": "Help me test my algorithm." }]
}
CSC 141 · Student allocationsIllustrative
01Student 014
32 / 100 calls
02Student 027
61 / 100 calls
03Student 038
18 / 100 calls
The same opportunity. An individual allocation. A faculty-set limit.
25Cloudflare-hosted models
for teaching and exploration
for teaching and exploration
1API endpoint
across your courses
across your courses
3Levels of visibility
student · course · department
student · course · department
A teaching resource, shared by the department
Broader access.
A more intentional investment.
Fund the learning environment, then let faculty shape how AI belongs in each course.
The instructor sets the rules
Choose available models, set per-student allocations, limit output length, and pause access when an assignment calls for independent work.
Make the spend visible
Allocate a monthly budget to each class. Track usage, reserve allowance before inference, and export a department-level report.
Build beyond a chat window
Bring model comparison, debugging, vision, tool calling, and application development into projects using a familiar API.
Built around the course
Your classroom.
Your AI policy.
A practical control panel for instructors, with student access connected to the course that funds it.
- Per-student daily requests and monthly Token / USD allocations
- Shared course budgets and model allowlists
- Key issuance, rotation, and access revocation
- Course pause, usage exports, and an administrative audit trail
Faculty workspaceSample data
CSC 141
ActiveIntroductory programming · example course
Monthly course allocation$120
$33.60 estimated usage
Student daily request limit100
Faculty controlled
| Student allocation | Monthly tokens | Access |
|---|---|---|
| Student 014 | 200,000 | Enabled |
| Student 027 | 400,000 | Custom quota |
| Student 038 | 200,000 | Enabled |
Room to experiment
One API. A range of capabilities.
Faculty enable the right models for each course. Students can compare approaches without changing their integration.
Everyday teaching
GLM-4.7-Flash
Dialogue, multilingual work, and tool calling.
Code & agents
GLM-5.3 / Flash
Explore reasoning and agentic programming.
Multimodal projects
Qwen 3.8 · Gemma 4
Combine images and text in student applications.
Advanced exploration
Kimi · DeepSeek V4
Long-context reasoning and tool workflows.
Model comparisons
GPT-OSS · Llama
Evaluate different open-model families.
Creative computing
FLUX image models
Instructor-enabled image generation.
Catalog verified September 11, 2026. Some models require paid Cloudflare access. Explore model IDs and published rates ↗
A defined ask. A measurable pilot.
Fund a semester
of possibility.
Start with an explicit student allocation, then review adoption, learning activities, and cost before expanding.
Illustrative funding request
$4,800
500 students · 4 months · $2 / student / month
Student AI allocations$4,000
Planning contingency$800
Proposed cost / student / semester$9.60
This is an editable planning scenario, not an approved budget, quote, or usage forecast. It reserves an AI allowance; hosting, support, and institutional requirements must be scoped for the pilot. Unused allocations do not imply an actual charge.
01 / Start deliberately
Recruit a small faculty cohort
Identify suitable assignments, set course policies, and issue student keys. Begin with a defined allocation.
02 / Measure the pilot
Review adoption and cost
Evaluate participation, aggregate usage, faculty feedback, and examples of student learning.
03 / Decide with evidence
Refine before expanding
Use the review to adjust allocations. Scope campus SSO, LMS integration, and institutional data review for a wider rollout.