Openrouter model routing

OpenRouter Model Routing

Overview

OpenRouter gives you access to 100+ models through one API. The key to cost efficiency is routing each request to the right model based on task complexity, required capabilities, cost budget, and latency requirements. This skill covers task-based routing, complexity classification, cost-aware selection, and OpenRouter's native routing features.

Prerequisites

• An OpenRouter API key exported as OPENROUTER_API_KEY — see the openrouter-install-auth skill for setup
• Python 3.8+ with the OpenAI SDK and requests (pip install openai requests)
• A rough inventory of your task mix (classification, summarization, code generation, deep reasoning, ...) to seed the TASK_ROUTING table
• Credits sized for the tiers you route to — the premium tier (openai/o1) runs $15/$60 per 1M tokens, 250x the budget tier

Instructions

1. Define your tiers per Task-Based Router: the MODELS dict (free → budget → mid → standard → premium) and the TASK_ROUTING map, then send requests through route_request(), which returns content, the serving model, tier, and token count.
2. When callers can't label tasks, switch to the Complexity-Based Auto-Router — classify_complexity() scores word count, code, reasoning, and math markers to pick a tier inside auto_route().
3. Add resilience per OpenRouter Native Routing: extra_body={"models": [...], "route": "fallback"} tries models in order, provider.order controls which provider serves, and the :floor variant picks the cheapest provider automatically.
4. Keep pricing current per Cost-Aware Router — get_model_pricing() pulls live per-1M rates from GET /api/v1/models, and cheapest_model_for_task() selects under context/tooling constraints.
5. Log every routing decision (task type, tier, model, cost) and tune per Error Handling and Enterprise Considerations — escalate the tier on quality regressions and cap per-request cost with max_tokens.

Task-Based Router

import os, re
from openai import OpenAI

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key=os.environ["OPENROUTER_API_KEY"],
    default_headers={"HTTP-Referer": "https://my-app.com", "X-Title": "my-app"},
)

# Model tiers by cost and capability
MODELS = {
    "free":    "google/gemma-2-9b-it:free",          # $0/0 — testing only
    "budget":  "meta-llama/llama-3.1-8b-instruct",   # $0.06/$0.06 per 1M
    "mid":     "openai/gpt-4o-mini",                  # $0.15/$0.60 per 1M
    "standard":"anthropic/claude-3.5-sonnet",         # $3/$15 per 1M
    "premium": "openai/o1",                           # $15/$60 per 1M
}

TASK_ROUTING = {
    "classification":  "budget",   # Simple label assignment
    "translation":     "mid",      # Moderate quality needed
    "summarization":   "mid",      # Good quality, cost-effective
    "code_generation": "standard", # Needs high accuracy
    "code_review":     "standard", # Needs reasoning
    "analysis":        "standard", # Complex reasoning
    "creative_writing":"standard", # Quality matters
    "deep_reasoning":  "premium",  # Multi-step logic
    "simple_qa":       "budget",   # Basic questions
    "chat":            "mid",      # General conversation
}

def route_request(task_type: str, messages: list[dict], **kwargs) -> dict:
    """Route to appropriate model based on task type."""
    tier = TASK_ROUTING.get(task_type, "mid")
    model = MODELS[tier]

    response = client.chat.completions.create(
        model=model, messages=messages, **kwargs
    )
    return {
        "content": response.choices[0].message.content,
        "model": response.model,
        "tier": tier,
        "tokens": response.usage.prompt_tokens + response.usage.completion_tokens,
    }

Complexity-Based Auto-Router

def classify_complexity(prompt: str) -> str:
    """Classify prompt complexity to select model tier.

    Simple heuristics -- replace with a trained classifier for production.
    """
    word_count = len(prompt.split())
    has_code = bool(re.search(r'```|def |function |class |import ', prompt))
    has_reasoning = bool(re.search(r'explain|analyze|compare|why|how does|trade.?off', prompt, re.I))
    has_math = bool(re.search(r'calculate|equation|formula|derive|proof', prompt, re.I))

    if has_math or (has_reasoning and has_code):
        return "premium"
    if has_code or has_reasoning or word_count > 500:
        return "standard"
    if word_count > 100:
        return "mid"
    return "budget"

def auto_route(messages: list[dict], **kwargs):
    """Automatically select model based on prompt complexity."""
    user_msg = next((m["content"] for m in reversed(messages) if m["role"] == "user"), "")
    tier = classify_complexity(user_msg)
    model = MODELS[tier]

    response = client.chat.completions.create(model=model, messages=messages, **kwargs)
    return response

OpenRouter Native Routing

# Route: "fallback" — try models in order until one succeeds
response = client.chat.completions.create(
    model="anthropic/claude-3.5-sonnet",
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=200,
    extra_body={
        "models": [
            "anthropic/claude-3.5-sonnet",
            "openai/gpt-4o",
            "openai/gpt-4o-mini",
        ],
        "route": "fallback",
    },
)

# Provider routing — control which provider serves a model
response = client.chat.completions.create(
    model="anthropic/claude-3.5-sonnet",
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=200,
    extra_body={
        "provider": {
            "order": ["Anthropic", "AWS Bedrock"],
            "allow_fallbacks": True,
        },
    },
)

# Model variant: ":floor" picks cheapest provider
response = client.chat.completions.create(
    model="anthropic/claude-3.5-sonnet:floor",
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=200,
)

Cost-Aware Router

import requests

def get_model_pricing() -> dict:
    """Fetch current pricing for cost-aware routing."""
    models = requests.get("https://openrouter.ai/api/v1/models").json()["data"]
    return {
        m["id"]: {
            "prompt": float(m["pricing"]["prompt"]) * 1_000_000,
            "completion": float(m["pricing"]["completion"]) * 1_000_000,
            "context": m["context_length"],
        }
        for m in models
    }

def cheapest_model_for_task(pricing: dict, min_context: int = 4096,
                             needs_tools: bool = False) -> str:
    """Find the cheapest model that meets requirements."""
    candidates = [
        (mid, p) for mid, p in pricing.items()
        if p["context"] >= min_context and p["prompt"] > 0  # Exclude free (unreliable)
    ]
    candidates.sort(key=lambda x: x[1]["prompt"] + x[1]["completion"])
    return candidates[0][0] if candidates else "openai/gpt-4o-mini"

Output

• Routed completion dicts from route_request(): the reply content, the actual model that served, the tier chosen, and total tokens consumed
• Router decision traces per request, e.g. [Router] Task=code -> Model=anthropic/claude-3.5-sonnet, giving you an audit trail to tune the routing table against
• A live pricing map from get_model_pricing() keyed by model ID: per-1M prompt/completion cost plus context length for cost-aware selection

Examples

The same router sends trivial and demanding prompts to opposite ends of the cost spectrum:

print(routed_completion("What is 2+2?"))
# [Router] Task=simple -> Model=google/gemma-2-9b-it:free

print(routed_completion("Write a Python function to merge two sorted lists."))
# [Router] Task=code -> Model=anthropic/claude-3.5-sonnet

The 4-word arithmetic prompt lands on the free tier while the code request escalates to Claude 3.5 Sonnet — the spread between those two decisions is where the cost savings live. More worked examples: references/examples.md.

Error Handling

Error	Cause	Fix
Wrong model selected	Classification too coarse	Add more task categories; test with diverse prompts
Model unavailable	Selected model temporarily down	Add fallback chain per tier
Cost overrun	Complex tasks routed to premium models	Set max_tokens and daily budget caps
Quality regression	Budget model can't handle task	Monitor output quality; escalate tier on poor results

Enterprise Considerations

• Start with manual task-type routing (explicit labels), then graduate to auto-classification
• Log every routing decision (task type, tier, model, cost) to tune the router over time
• Use OpenRouter's :floor variant to automatically get the cheapest provider for any model
• Set max_tokens on every request to cap per-request cost regardless of model tier
• A/B test routing rules: send 10% of traffic to a different tier and compare quality metrics
• Combine with fallback chains so each tier has backup models

References

• Examples | Errors
• Model Routing | Provider Routing