zeroclaw-labs/zeroclaw · error · anyhow::Error

Telnyx API error ({}): {}

Error message

Telnyx API error ({}): {}

What it means

chat_with_system sent a chat completion to Telnyx and got a non-success status; the body is passed through sanitize_api_error to strip credentials before being embedded. The status code is included, so 401/400/429/500 each point at a distinct fix.

Source

Thrown at crates/zeroclaw-providers/src/telnyx.rs:206

            model: model.to_string(),
            messages,
            temperature,
        };

        let response = self
            .client
            .post(self.chat_url())
            .header("Authorization", format!("Bearer {}", api_key))
            .header("Content-Type", "application/json")
            .json(&request)
            .send()
            .await?;

        if !response.status().is_success() {
            let status = response.status();
            let error = response.text().await?;
            let sanitized = super::sanitize_api_error(&error);
            anyhow::bail!("Telnyx API error ({}): {}", status, sanitized);
        }

        let chat_response: ChatResponse = response.json().await?;

        chat_response
            .choices
            .into_iter()
            .next()
            .map(|c| c.message.content)
            .ok_or_else(|| {
                ::zeroclaw_log::record!(
                    ERROR,
                    ::zeroclaw_log::Event::new(module_path!(), ::zeroclaw_log::Action::Fail)
                        .with_outcome(::zeroclaw_log::EventOutcome::Failure),
                    "telnyx: empty choices in response"
                );
                anyhow::Error::msg("No response from Telnyx")
            })

View on GitHub (pinned to 88bb9c8533)

Solutions

1. Read the status in the message: 401 → fix the API key; 400 → fix model name/parameters.
2. 429 → back off and retry with exponential delay, or lower request rate.
3. Confirm the model id in the provider entry is one Telnyx's models endpoint actually lists.
4. 5xx → check Telnyx status page and retry later.

Example fix

# before — model id not offered by Telnyx
[providers.models.telnyx.main]
model = "gpt-4o"

# after — an id that appears in the Telnyx models list
[providers.models.telnyx.main]
model = "meta-llama/Llama-3.1-70b-instruct"

Parse the status from the message: retry 429 and 5xx with exponential backoff + jitter (cap ~3 retries); fail immediately on 400/401 — they are deterministic config errors. Log the sanitized body verbatim for diagnosis.

When it happens

Trigger: 401 from a bad key; 400 from an unsupported model name or invalid parameters for a single system-prompt turn; 429 rate limiting; 5xx provider errors.

Common situations: Key rotated but config still holds the old one; model id not enabled for the account; bursty traffic hitting Telnyx quotas; Telnyx incident.

Related errors

• Failed to list Telnyx models: {}
• Ollama API error ({}): {}. Is Ollama running? (brew install
• channel does not support room creation
• elicitation returned unknown choice const: {s}
• purge_namespace not supported by this memory backend

AI-assisted analysis of zeroclaw-labs/zeroclaw@88bb9c8533 (2026-08-23). Data as JSON: /api/errors/5d81abab0276f379. Report an issue: GitHub.