Update, Nov 12, 2026: CodeRouter's pricing has changed since this post. It is now pay as you go — model list price with no markup from a prepaid balance, plus a 5.5% payment fee when you buy credit, no monthly fee, with smart routing or Direct mode (you name the exact model) on the same account — or Team at $19 per developer a month plus usage at list price. The monthly plans and the separate Direct plan described below are no longer sold; accounts already on them keep their terms. The rest of the post is as published. Current terms: pricing.
TL;DR — Our pitch is "smart routing": each request gets classified by intent (planning, implementing, debugging, testing, documenting) and routed to the cheapest capable model. Opus only when truly needed. But last week's audit caught us: 80%+ of traffic was bypassing the router because client defaults (Claude Code / Cursor / Codex) hardcoded
model: "claude-opus-4-7"and we honored it. 12 hours of fixes later, real data (2,094 requests): per-request cost dropped from $0.19 to $0.122 (-36%), Opus share of spend from 80%+ to 45%, and spend 91% below an all-Opus estimate (it was 60% before the fix). Here's the process and the lessons.
What was measured: CodeRouter's own production traffic in May 2026 (mostly our own team's development work), priced at provider list prices. The all-Opus figures are estimates: the same tokens priced at Opus list price. Routing, models and plans have changed since — see the update at the end.
1. The embarrassing thing we found
We sell smart routing. Each request should get classified by intent — planning, implementing, debugging, testing, documenting — and routed to the cheapest model that can handle it. In theory: Opus only for the heavy reasoning, V4-Pro for implementation, V4-Flash for tests and docs. Most everyday coding requests should NOT use Opus.
But last week's product audit, on 5,000 real requests:
Requests where the router actually ran: 15%
Requests that bypassed the router: 85%
Why? Users' clients (Claude Code / Cursor / Codex) ship with default configs that hardcode model: "claude-opus-4-7" — and we dutifully forwarded those to Opus. Smart routing got skipped entirely.
This was a product-level embarrassment: the core value we were selling didn't get a chance to fire on most client default configurations.
2. Our call: force auto on Smart-tier plans
The fix was a product positioning decision more than a technical one. Two options:
A. Educate users. Email everyone, ask them to flip model: "auto" in their config.
B. Override the model field server-side. Smart-tier plans always go through smart routing, regardless of what the client sends.
We picked B. Reason: smart routing IS the product. If a customer paid for the Smart tier and their client default bypassed the routing, that's our responsibility, not the user's.
But B has a precondition: give users who genuinely want manual model control a way out. We already had the Direct plan (pay-as-you-go: you pick the model and pay provider list price +5%). It became the explicit-mode escape hatch.
So the product positioning sharpened into two distinct lanes:
- Smart plans (Starter / Solo / Pro / Team): subscription, smart routing, whatever model you send is rewritten to auto
- Direct plan: pay-as-you-go, the only plan where your
model:choice is honored
3. The data, 12 hours later
After the fix, 12 hours of real production traffic — 2,094 requests:
| Phase | Share | Routed to (top 3) |
|---|---|---|
| debug | 41.7% | Opus 44% + Sonnet 22% + GPT-5.5 18% + V4-Pro 15% |
| implement | 32.2% | V4-Flash 62% + V4-Pro 16% + GPT-5.4 12% |
| test | 17.2% | V4-Flash 47% + Sonnet 34% + V4-Pro 19% |
| plan | 5.1% | V4-Pro 77% + Sonnet 8% |
| document | 2.5% | V4-Flash 94% |
| small_edit | 0.9% | gpt-5-mini 100% |
Each phase landed on the right model mix. Opus is still the first pick for genuinely reasoning-heavy debug work, but everywhere else it stepped aside for cheaper models.
4. The cost numbers
12-hour actual spend: $254.89
12-hour all-Opus estimate: $2,819.63
Below the all-Opus estimate: 91.0%
We're not annualizing this. Twelve hours is a short window, and the all-Opus figure is an estimate, not a bill. The cleaner before/after comparison is the per-request cost: $0.19 before the fix, $0.122 after (-36%).
5. Three counterintuitive findings
A. Planning doesn't need Opus
We'd assumed planning required Opus-tier reasoning depth. Reality: of 107 plan-phase requests in 12 hours, 77% routed to V4-Pro and 0% to Opus.
Reason: V4-Pro and Opus both score top-tier on plan-quality benchmarks, but V4-Pro costs ~10× less. The balanced strategy picks V4-Pro on quality/cost ratio. User feedback didn't show any quality regression.
Takeaway: "Opus is mandatory for planning" was an assumption, not a fact.
B. One cheap model carried 32% of traffic for $2
V4-Flash (DeepSeek's flash tier) handled 659 requests in 12 hours for a total cost of $2.13. Same volume on Opus would have been $300+.
It's especially good on long-context Chinese-language sessions, where 98% prompt-cache hit rates make per-token cost essentially free.
C. Most "routing problems" are actually "config problems"
We'd spent weeks improving the routing classifier — multilingual support, tool-call inference, multi-turn fallbacks. But what actually moved savings from 60% to 91% wasn't the algorithm. It was:
- Forcing client requests through the routing pipeline (the silent rewrite)
- A sensible default for ambiguous cases (default to "implement" → cheap models)
The classifier was already good enough. The problem was that most requests never got classified.
6. Three lessons
- Product defaults beat algorithm precision. A clever router doesn't matter if client defaults bypass it. If you're selling routing, you have to actually route.
- Build a real escape hatch for the "I want control" segment. Smart plans = forced auto, Direct plan = explicit selection. Two clear lanes, no fuzzy middle.
- Public data is the moat. Most LLM resellers don't publish real audit data because their differentiation is just "cheaper API." Our differentiation is the routing logic itself, which transparency only strengthens.
7. What this means for you
If your team uses Cursor / Claude Code / Codex and your monthly bill is over $1,000, there's a good chance you're stuck in the same trap — every request defaulting to Opus or GPT-5.5 because that's what your client config says.
How to check: pull your last 30 days of usage. Look at the share of cost going to Opus (or GPT-5.5 / GPT-5.4). If it's over 60%, and your team isn't full of architects doing nothing but planning, the excess is likely overspend.
CodeRouter's free trial gives you $1 of usage over 7 days with no card, $5 once you add a card (it isn't charged). Paid plans include model usage in dollars at list price — Pro is $99/month with $85 included (Opus counts toward it); teams use the Team plan at $19 per developer plus usage at list price +5%. Details on /pricing. Try it for a week and compare the per-request cost on your dashboard with what you were paying.
Update 2026-11-04: Routing has changed since this post. Routing v3 uses a small judgment model (Jev) plus cost rules: each request starts on the cheapest tier that can handle it — Claude Haiku 5.5 (or DeepSeek Chat), then Kimi K2.6, Claude Sonnet 5.5 and Claude Opus 5.5 — and moves up only when tests or errors show a failure. The models in the table above (V4-Flash, V4-Pro, GPT-5.5) are what routing used in May 2026. For a measured result on fixed tasks: in TokenWatt's 176-task coding benchmark (TokenWatt is a measurement project from the same team as CodeRouter), routing solved 96–100% of tasks for 56–70% less than sending every request to Opus, across runs.
All data here comes from real production audits on our own infrastructure. We publish honest retrospectives because they're more persuasive than marketing copy.