Docs

Mistral AI

Ingest Mistral traces into AITracer via the SDK or API.

Agent Lab: Run local Ollama agents in the dashboard — no SDK required (Quick Start). This guide is for ingesting traces from this provider via the SDK or API.

Use the SDK or API to store Mistral chat completions in AITracer — same Traces history as local Ollama runs.

Each ingested record typically includes:

messages / response text
model id (mistral-large-latest, etc.)
prompt_tokens, completion_tokens, total_tokens
latency and finish reason

Mistral integration workflow

Rendering diagram...

Connecting Mistral (coach / cloud execution)

Settings → Provider Connections → Mistral AI
Paste your Mistral API key

Use SDK ingest from application code for production traffic regardless of dashboard settings.

What is captured

Full conversation messages (or summarized prompt/response)
Token usage from the Mistral response
Generation parameters (temperature, max tokens) when provided in parameters

Cost tracking

Token-based cost estimates appear when metrics are included on ingest. See Cost Intelligence for rollups.

Optional advanced

Governance Engine, Verification Layer, and Audit Vault apply when enabled on your deployment.