Taylor Hawkes
Founder and engineer, Inference APIs.
Taylor builds and operates Inference APIs: the gateway, the model catalog, the playgrounds and the billing. Everything in the reference and documentation is written and verified against the live API and the providers' own documentation, with the verification date shown on each page.
Elsewhere: GitHub. Reach the company at support@inferenceapis.com.
Writing
- What Codex CLI sends to your provider · 2026-09-17
We captured every request Codex CLI 0.154.0 makes when pointed at a custom provider: a 17,000-character system prompt on every turn, 17,500 characters of tool definitions, a stable installation id, and two request shapes that only OpenAI understands. Here is what is in them, why MCP tools and sub-agents break on open models, and what a provider can do about it. - Whisper Large v3 tested on 83 languages · 2026-09-17
Whisper lists 100 languages. We sent 30 public test clips per language through our API and scored every transcript. Around a third of languages are good or excellent, a third are not usable, a few come back empty, and auto-detection confuses more languages than you would expect. - Measured: latency and speed of nine open chat models · 2026-09-16
We timed every chat model on the public endpoint with a fixed prompt: time to the first token, how long each model thinks before the visible answer starts, and output tokens per second. Includes a correction to the figures we first published. - Groq retired llama-3.3-70b-versatile: what to do · 2026-09-16
On 16 August 2026 Groq shut down llama-3.3-70b-versatile and llama-3.1-8b-instant for free and developer tiers. A month later, hundreds of thousands of code files still pin the id. What the error looks like, who it hit, and the fix ladder. - How we price inference: cost plus a fixed margin · 2026-09-16
Every model is priced at about 30% over what it costs us to serve, and the rate is on the model page. What that means in practice, and why we do not have tiers.
Also maintained: the provider error reference, the free-tier rate limits tracker and the model deprecations tracker, each re-verified against provider documentation on the date shown on the page.