Blog Model Guides
DeepSeek V4 Pro and Flash Production Routing Checklist
Make the route observable before making it automatic.
AIWave prices checked 2026-11-05 UTC against the live pricing response and active default-group routes. The old price and capacity examples on this URL were not current evidence. Check the live rate card before using a number in production.
| Exact model ID | Fresh input | Cache-hit input | Output |
|---|---|---|---|
deepseek-v4-flash | $0.30 | $0.006 | $1.20 |
deepseek-v4-pro | $1.32 | $0.044 | $3.96 |
Before rollout
- Check the exact Pro and Flash model IDs in the live rate card and route table.
- Prepare a test set with pass/fail criteria and a maximum retry budget.
- Record request ID, effective model and group, response status, usage, elapsed time, and final charge.
- Test tool side effects separately; a retry must not duplicate an external action.
During the staged rollout
Start with a bounded sample of non-critical tasks. Compare completed-task cost, acceptance rate, and latency by model. Watch error and fallback rates. If the route cannot prove the selected model or ledger reconciliation, pause the automated switch and return to the last verified policy.
Price and capacity are separate decisions
The checked table provides default-group text-token rates. It does not promise context size, concurrency, peak discounts, or a provider capacity limit. Verify those properties with their own primary evidence and controlled requests before encoding them as routing rules.
For an executable first request, use the quickstart. Recheck the effective account group and request settlement before treating an estimate as a bill.