This guide uses source checks from Sep 24, 2026. Provider and gateway prices can change; preserve the checked date with every forecast.
Why This Topic Matters Now
The Sep 23 report retained `glm api` intent, but a model name is not a contract. Tool calling introduces a second state machine: the model proposes an action, the application validates it, a tool runs, and the model may continue or stop. Structured output adds another gate. If those transitions are not tested, a response that looks correct can still trigger the wrong application behavior.
This guide gives Tier 1 and Tier 2 API teams a compact contract suite for GLM tool calls and structured responses. It focuses on schema boundaries, malformed arguments, tool failures, finish reasons, and source-dated route evidence. The comparison is operational rather than a leaderboard: the approved model is the one that passes the workload's policy and can be rolled back cleanly.
Source Facts Checked Today
AIWave /api/pricing was checked from production on Sep 24, 2026 and returned HTTP 200, success=true, 73 live route rows, pricing_version a42d372ccf0b5dd13ecf71203521f9d2, auto_groups=['default'], group_ratio default=1 and vip=0.9. The public /api/v1/pricing endpoint returned HTTP 200 with 56 dated USD rows, pricing_version 83f77abde81ee3a096a672ed959ccc096f5d37a45c177ae8e03229456b5415a5, checked=2026-09-10, and updated_at=2026-09-18. Use the live response for route availability and the dated JSON for a forecast; they are not one interchangeable rate table.
The live AIWave route response checked on Sep 24, 2026 included GLM route rows for OpenAI-compatible use. The live response establishes current gateway availability metadata; it is not a provider-direct invoice or proof that every provider-specific tool feature is exposed.
The dated AIWave public pricing JSON checked in this run lists glm-5 at $1.55 input, $0.40000075 cache-hit input, and $4.96 output per 1M tokens, effective 2026-08-27. It lists glm-5-turbo at $1.80 input, $0.4800006 cache-hit input, and $5.40 output per 1M tokens, also effective 2026-08-27. Keep these rows separate from any Z.AI provider table.
Z.AI pricing and quick-start pages checked on Sep 24, 2026 provide provider documentation and onboarding context. Because provider pages can change, use them as dated source links and rerun the contract tests before changing a production model policy.
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.
| Test stage | Bad outcome | Required assertion |
|---|---|---|
| Argument generation | Unknown or unsafe field appears | Schema and allowlist reject it |
| Tool dispatch | Wrong tool receives the call | Tool name and version match |
| Tool result | Failure is treated as success | Status and bounded error are explicit |
| Continuation | Model loops after a tool error | Maximum tool turns and stop reason |
| Structured output | Valid JSON has invalid meaning | Schema plus semantic validator |
| Rollback | Model change breaks a contract | Previous model ID and fixture set |
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.
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY_HERE",
base_url="https://aiwave.live/v1",
)
allowed_tools = {"lookup_release": {"version": "v2"}}
result = client.chat.completions.create(
model="glm-5",
messages=[{"role": "user", "content": "Return one release risk as JSON."}],
tools=[{"type": "function", "function": {"name": "lookup_release",
"description": "Read a redacted release record", "parameters": {"type": "object",
"properties": {"release_id": {"type": "string"}},
"required": ["release_id"], "additionalProperties": False}}}],
temperature=0.0,
max_tokens=260,
)
print({"finish": result.choices[0].finish_reason, "usage": result.usage,
"allowed_tools": list(allowed_tools)})
Turn the Query Into a Contract
For GLM tool-calling and structured-output tests, define the request shape, model ID, data class, output ceiling, timeout, retry ceiling, owner, and source date before the first trial. A short contract gives engineering, security, and finance the same object to review when a provider changes a route or billing field.
Separate Live Routes From Dated Rates
The live AIWave pricing response answers which route rows and endpoint types are available at check time. The public pricing JSON is a dated USD snapshot for forecasting. Store both URLs, versions, checked dates, model IDs, and account-group context instead of presenting a volatile source as a permanent quote.
Use a Small Acceptance Set
A useful canary covers a normal request, a malformed request, a repeated prefix, a long output, a disconnect, and a deliberate stop condition. Record request ID, model ID, status, token usage, finish reason, retry count, and reviewer outcome. This turns a search result into evidence that can survive a route update.
Keep Data and Credentials Bounded
OpenAI-compatible clients reduce integration work, but they do not choose the right data boundary. Keep the credential server-side, use a visible placeholder in examples, redact fixtures, and attach a data-class decision to every route policy. Do not let a feature flag or model alias silently widen what crosses the API.
Make Recovery Observable
Retry only failures that are safe to retry and cap every fallback. Preserve the original request ID, mark the stop reason, and distinguish provider errors from client validation, policy rejection, and budget stops. Silent loops hide both reliability failures and billing variance.
Use AIWave's Evidence Layer
Use the Models docs, Chat Completions docs, live pricing API, and Trust. Recheck the live route table before rollout, the dated pricing JSON before a budget review, the status page before a launch window, and the trust page before procurement. Keep each checked date visible in the record.
Release Gate
Promotion is ready when the provider source is dated, the AIWave route is rechecked, the acceptance set passes, the billing fields are understood, and a named owner can stop or reverse the change. If a field is unknown, label the work as a trial rather than production.