zed-industries/zed · error

Failed to connect to DeepSeek API: {} {}

Error message

Failed to connect to DeepSeek API: {} {}

What it means

DeepSeek chat-completion client, non-streaming branch: when the HTTP response status is not successful, the body is read to string and the error bails with the status code plus the provider's raw error body (deepseek.rs:336). The body usually contains DeepSeek's JSON error object naming the actual cause.

Source

Thrown at crates/deepseek/src/deepseek.rs:336

            .lines()
            .filter_map(|line| async move {
                match line {
                    Ok(line) => {
                        let line = line.strip_prefix("data: ")?;
                        if line == "[DONE]" {
                            None
                        } else {
                            Some(serde_json::from_str(line).map_err(Into::into))
                        }
                    }
                    Err(error) => Some(Err(anyhow!(error))),
                }
            })
            .boxed())
    } else {
        let mut body = String::new();
        response.body_mut().read_to_string(&mut body).await?;
        anyhow::bail!(
            "Failed to connect to DeepSeek API: {} {}",
            response.status(),
            body,
        );
    }
}

View on GitHub (pinned to f4178619ac)

Solutions

1. Match the status: 401 → fix the DeepSeek API key in Zed's settings; 402/429 → add balance or back off and retry; 404 → correct the model id; 5xx → retry with backoff later.
2. Read the Body text in the message — DeepSeek returns a JSON error message describing the exact problem.
3. Reproduce outside Zed with curl using the same key/model to confirm whether it is config or provider-side.
4. If behind a proxy, ensure it forwards the Authorization header and does not rewrite the request.

# confirm key/model/network before relying on the integration
curl -sS -o /dev/null -w '%{http_code}' https://api.deepseek.com/chat/completions \
  -H "Authorization: Bearer $DEEPSEEK_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"model":"deepseek-chat","messages":[{"role":"user","content":"ping"}]}'
# expect 200; 401 -> key, 404 -> model, 429 -> backoff

Parse the status out of the error; retry 429/5xx with exponential backoff and jitter, fail fast on 400/401/404 with the provider's body message shown to the user.

When it happens

Trigger: POST to DeepSeek's API returning 401 (invalid API key), 402/429 (insufficient balance / rate limit), 404 (unknown model id), 400 (malformed request), or 5xx (provider incident); anything other than a 2xx on a non-streaming completion.

Common situations: Expired or rotated API key; wrong model string after DeepSeek renames models; hitting RPM/TPM limits under load; regional network blocks; provider outages.

Related errors

• Failed to connect to Mistral API: {} {}
• Request failed with status: {status:?} Body: {body}
• Request failed with status: {:?} Body: {}
• custom server error: {} - {}
• status error {}, response: {text:?}

AI-assisted analysis of zed-industries/zed@f4178619ac (2026-08-20). Data as JSON: /api/errors/3b03ae0323c44956. Report an issue: GitHub.