AIWave Pricing JSON vs Live Route Table for API Budgets

Why This Topic Matters Now

The Sep 13 keyword report still shows a Tier 1 brand and site-search cluster: `aiwave.live`, `site:aiwave.live`, `aiwaveblog`, and `aiwave api`. Those are not broad research terms. They are evidence-seeking queries from developers and evaluators who want a page, a JSON endpoint, or a route table they can inspect. The right answer is not another generic platform summary; it is a source-dated guide to the two pricing surfaces that matter before a budget forecast.

AIWave now exposes both a public pricing JSON snapshot and a live operational route table. They should work together, but they should not be treated as the same artifact. The public JSON gives a stable, dated base-rate reference that can be cited in docs, procurement notes, and model-selection workbooks. The live route table tells engineers what is enabled now, which pricing version is active, and which group multipliers apply to the account or key used in the actual request.

Source Facts Checked Today

AIWave /api/pricing checked on Sep 14, 2026 returned success=true, 64 live rows, pricing_version 5a90f2b86c08bd983a9a2e6d66c255f4eaef9c4bc934386d2b6ae84ef0ff1f1f, auto_groups=['default'], and group_ratio default=1 and vip=0.9. The static /api/v1/pricing snapshot checked the same day reported checked=2026-09-10, 64 rows, pricing_version 8c7a0c0b30661ccbc13d142cb54d1e4ae445fe774b2c6fa501080db97c7a3e56, and notes that dated base rates are adjusted by the effective account group. Selected static base rows per 1M text-token units were DeepSeek Flash at $0.70 input, $0.0233 cache-hit input, and $2.10 output; DeepSeek V4 Flash at $0.638 input, $0.0202884 cache-hit input, and $1.914 output; DeepSeek V4 Pro at $1.914 input, $0.0637362 cache-hit input, and $5.742 output; GLM-4.5 at $0.6975 input, $0.1800003375 cache-hit input, and $2.1699999225 output; Qwen3 Max at $1.5621977891181764 input and $6.248791156472706 output; Kimi K3 at $4.50 input, $0.90 cache-hit input, and $22.50 output; and Moonshot v1 128K at $1.80 input and $4.50 output. VIP-key estimates multiply the same base rows by 0.9.

QwenCloud pricing documentation checked for this run describes different billing modes for text, image, video, speech, embedding, reranking, and built-in tools, with batch and context-cache concepts called out separately. That supports a feature-family budget rather than a single blended AI line item.

Z.AI pricing documentation checked for this run lists model prices per 1M tokens and separate built-in tool pricing, including Web Search at $0.01 per use. That official-provider context is useful for GLM comparisons, but provider rows must remain separate from AIWave gateway rows unless the gateway surface publishes the same route and date.

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 requests

BASE = "https://aiwave.live"

public_pricing = requests.get(f"{BASE}/api/v1/pricing", timeout=20).json()
live_pricing = requests.get(f"{BASE}/api/pricing", timeout=20).json()

snapshot = {
    "api_key": "YOUR_API_KEY_HERE",
    "public_checked": public_pricing["checked"],
    "public_pricing_version": public_pricing["pricing_version"],
    "live_pricing_version": live_pricing["data"][0]["pricing_version"],
    "group_ratio": live_pricing.get("group_ratio"),
    "model_count": len(live_pricing["data"]),
}

print(snapshot)

Use Public JSON for Stable Citations

The public JSON endpoint is the right artifact when an evaluator wants a dated row that can be pasted into a workbook. It contains currency, unit, source, checked date, updated date, pricing_version, and model rows with provider names and token classes. That makes it easier to review a forecast two weeks later. It also keeps the buyer from depending on a screenshot or a stale memory of a pricing page.

Use the Live Table Before a Run

The live route table is the right artifact immediately before a production canary, release, or funding decision. It shows operational rows, route names, enabled groups, supported endpoint types, group_ratio, and the active pricing_version for the table. A team should capture it before changing traffic because route availability and group multipliers are operational facts, not marketing copy.

Keep Group Math Explicit

The live Sep 14 check reports default=1 and vip=0.9. That means a base row such as DeepSeek Flash at $0.70 input and $2.10 output remains the default public row, while a VIP-key estimate applies the 0.9 multiplier. The forecast should show both base and applied rows, and it should label whether the key was default or VIP. Without that field, month-end variance review turns into guesswork.

Separate Provider Context

Provider documentation explains source mechanics: token classes, cache concepts, media billing, tool fees, and model families. Gateway evidence explains what happened through AIWave. A QwenCloud tool fee or a Z.AI Web Search row should not be pasted into an AIWave forecast unless the actual route and usage receipt support it. Keep provider context in a comparison tab and gateway facts in the route tab.

Build a Pricing Evidence Pack

A useful evidence pack has six small files: public pricing JSON, live route table, status endpoint result, selected provider docs URL list, redacted request receipt, and release note. The pack does not need customer data or prompts. It needs enough structure for engineering, finance, and procurement to agree on route name, date, token class, group multiplier, and fallback policy.

Internal Links for Search Evaluators

This article should connect brand-search users to Pricing, Pricing JSON, Status, Models docs, Chat Completions, and Trust. That path answers the common site-search question: which AIWave page should I inspect first?

Procurement Review

Procurement should ask for the checked date, pricing_version, route string, account or key group, model input row, cache-hit row if present, output row, tool or media fee class, and final usage receipt. If the team cannot show both the pre-run pricing source and the post-run receipt, the budget review should remain provisional. This is a practical standard, not ceremony.

Final Checklist

Before scaling a workload, download the public JSON, capture the live route table, record the group multiplier, run a redacted request, store the usage receipt, and link the provider docs used for context. Repeat the capture when the pricing_version changes or when the key group changes. That small habit keeps pricing drift visible and keeps Tier 1 evaluators from relying on stale numbers.