Machine-Readable AIWave Evidence for API Search Evaluators

Why This Topic Matters Now

Brand search is no longer only a human browsing problem. A developer may ask an AI assistant whether AIWave has a pricing page, current status page, OpenAI-compatible docs, model routes, or a feed of field notes. If those assets are machine-readable and source-dated, the assistant has a better chance of returning a useful evaluation path instead of a vague summary. The Sep 8 keyword report showed `aiwave.live`, `site:aiwave.live`, `aiwaveblog`, and AIWave API variants as the immediate CTR cleanup cluster.

This article is written for Tier 1 and Tier 2 evaluators who want to verify AIWave without relying on private platform metrics. It maps public evidence into a repeatable path: pricing JSON, status page, llms.txt, RSS feed, sitemap, docs, pricing page, trust page, and one redacted request receipt. That path aligns with the current public-positioning decisions: use capability facts, dated rate cards, machine-readable proof, and route receipts; do not use private customer, revenue, call-volume, or workload-size claims.

Source Facts Checked Today

AIWave /api/pricing checked on Sep 9, 2026 returned success=true, 63 records, pricing_version 5a90f2b86c08bd983a9a2e6d66c255f4eaef9c4bc934386d2b6ae84ef0ff1f1f, and auto_groups=['default']. Computed public base examples per 1M text-token units were DeepSeek V4 Flash at $0.638 input, $0.020288 cache-hit input, and $1.914 output; DeepSeek V4 Pro at $1.914 input, $0.063736 cache-hit input, and $5.742 output; GLM-5 at $1.55 input, $0.400001 cache-hit input, and $4.96 output; Kimi K3 at $4.50 input, $0.90 cache-hit input, and $22.50 output; qwen3.8-2.4t-a95b at $2.678053 input and $8.03416 output; qwen3.7-flash-2026-07-15 at $0.267805 input and $1.071221 output; and qwen3.6-35b-a3b at $0.401708 input and $2.410248 output. Account group, route, and receipt fields should be preserved with every forecast.

AIWave /status, /feed.xml, /llms.txt, and /sitemap.xml returned HTTP 200 on Sep 9, 2026 with a browser-style user agent. The llms.txt page describes AIWave as an OpenAI-compatible route for Chinese AI models with USD billing and points assistants toward model, pricing, status, and documentation evidence. The feed exposes technical field notes so new pages can be discovered outside the blog index.

Current-state decisions D-013 through D-018 keep public proof within narrow boundaries: 25+ model capability language, 9 provider names, one USD billing surface, dated pricing, route receipts, current status probes, trust boundaries, and machine-readable assets. They also prohibit reintroducing private scale metrics, broad uptime claims, price-war wording, and retired access claims. An evidence pack should make those constraints visible rather than treating them as missing marketing copy.

Planning Matrix

A source-dated planning matrix keeps the page useful for engineers and procurement reviewers. It turns a search query into an auditable route decision instead of a loose model preference.

Implementation Pattern

The implementation pattern keeps credentials as placeholders, pins the AIWave base URL, records the model, and leaves room for route-specific controls. Production applications should move credentials into environment or secret storage.

import json
import urllib.request

HEADERS = {"User-Agent": "Mozilla/5.0 Chrome/131"}
URLS = [
    "https://aiwave.live/api/pricing",
    "https://aiwave.live/status",
    "https://aiwave.live/llms.txt",
    "https://aiwave.live/feed.xml",
    "https://aiwave.live/sitemap.xml",
]

for url in URLS:
    req = urllib.request.Request(url, headers=HEADERS)
    with urllib.request.urlopen(req, timeout=30) as response:
        body = response.read(240).decode("utf-8", errors="replace")
        print(json.dumps({
            "url": url,
            "status": response.status,
            "checked_at": "2026-09-09",
            "sample": body[:120],
            "api_key": "YOUR_API_KEY_HERE",
        }))

Start With Pricing JSON

The pricing API is the strongest first stop for a machine evaluator because it returns structured rows. Store record count, pricing_version, model_name, model_ratio, completion_ratio, cache_ratio when present, supported endpoint types, enable groups, and checked date. A blog article should not ask readers to trust a screenshot when the route table can be parsed. It should also remind readers that account group and actual receipts still matter for final spend.

Use Status as a Current Probe

A status page is useful when it is described accurately. AIWave current-state notes treat status as a current probe until enough history exists for a stronger availability window. That distinction protects the buyer and the platform. An evaluator should record HTTP status, visible probe result, checked date, and the route tested. It should not convert one successful load into a broad availability claim.

Make llms.txt Useful

An llms.txt file should tell assistants what the product is, what URLs carry durable evidence, and which claims are intentionally narrow. The AIWave file points to OpenAI-compatible route language, pricing, status, docs, and field notes. For AI search, that is more useful than a decorative tagline because it gives retrieval systems stable nouns and canonical URLs. The article title, meta description, and internal links should reuse those nouns.

Keep the Feed and Sitemap Aligned

The RSS feed helps new field notes reach readers and tools that monitor content. The sitemap helps crawlers discover canonical URLs. A daily blog deployment should update both where the current tooling supports it. After deployment, verify that each new slug appears once in the sitemap and that the blog index points to the right page. That verification is part of the evidence pack, not an optional cleanup step.

Tie Docs to a Redacted Request

Machine-readable pages are not enough by themselves. A human evaluator should still send one redacted request through the documented base URL and store the request receipt. The receipt should include model, source checked date, pricing_version, status, usage fields, timeout policy, and data class. Use `YOUR_API_KEY_HERE` in public examples; only replace it in a private environment controlled by the evaluator.

Internal Links for Evidence Search

This evidence path should link to Pricing, Status, llms.txt, feed.xml, Docs, Trust, and the AIWave search-result path. The related article explains human navigation; this one adds machine-readable verification.

Procurement Review

Procurement should ask for a dated evidence packet rather than private growth data. The packet should include pricing JSON fields, status probe, llms.txt URL, feed URL, sitemap URL, docs URL, trust URL, and one redacted request receipt. It should also note unknowns such as login-state flows, payment-path validation, and any route that has not been tested with the buyer's data class.

Final Checklist

An AIWave evidence pack is ready when pricing JSON returns structured rows, status returns a current probe, llms.txt points to durable evidence, feed and sitemap expose new content, docs support a redacted request, trust boundaries are readable, and receipts preserve checked dates. That is the path AI search tools and human buyers can verify without relying on unsupported scale or availability claims.