Text model
Phi-3.5-MoE-instruct pricing
Azure AI Foundry charges $0.16 per million input tokens and $0.64 per million output tokens for Phi-3.5-MoE-instruct. Below: every rate that applies, what common workloads cost, and cheaper options.
- Input
- $0.16/ 1M
- Output
- $0.64/ 1M
- Context
- 128Ktokens
Retired 2025-08-30. npx @veqen/sdk scan finds every place your code uses it.
All rates
| Tier | Input | Output | Cache read | Cache write |
|---|---|---|---|---|
| Standard | $0.16 | $0.64 | — | — |
USD per million tokens. Long-context tiers apply when the whole prompt, cached tokens included, passes the threshold.
Work it out
What common workloads cost
| Workload | Per call | Cost per call | Calls a month | Per month |
|---|---|---|---|---|
| Chat reply | 1,500 input and 400 output tokens | $0.000496 | 100,000 | $49.60 |
| RAG answer, cached context | 2,000 fresh + 4,000 cached input, 500 output | $0.00128 | 100,000 | $128.00 |
| Agent step | 4,000 fresh + 8,000 cached input, 800 output | $0.00243 | 250,000 | $608.00 |
| Batch summary | 3,000 input and 300 output tokens via the Batch API | — | 1,000,000 | — |
Planning a mix of models? Simulate a workload with Phi-3.5-MoE-instruct, alongside anything else you run.
Cheaper options
Current models that cost at least 15% less for a typical chat reply. Cheaper isn't always equivalent: test quality on your own prompts, then use veqen diff to see the impact at your volume.
| Model | Chat reply | Saving |
|---|---|---|
| azure-ai/gpt-6-luna | $0.00035 | 29% |
| azure-ai/FW-DeepSeek-V4-Flash | $0.000349 | 30% |
| bedrock/anthropic.claude-haiku-5-5 | $0.00042 | 15% |
| cloudflare/@cf/openai/gpt-oss-20b | $0.00042 | 15% |
Price history
No price changes since Veqen started tracking this model.
Compare
Last verified with price book 2026-10-22. Sources: Azure AI Foundry pricing and LiteLLM's MIT-licensed price file. Track every call to this model, per customer or project: install the SDK.