Simon Willison updates ttok token counter with model listing
Source published ·Modelwire updated
Original coverage: Simon Willison ↗·How Modelwire adds context
The development
Simon Willison released ttok 0.4, a CLI utility for token counting that wraps OpenAI's open-source tiktoken library. The update adds a model-listing command and modernizes the codebase after two years of dormancy. The tool's integration with uvx makes token accounting accessible via simple piping, addressing a practical friction point for developers working with LLM APIs. For teams managing prompt costs and context windows, this represents a small but useful refresh in the token-accounting toolchain that underpins LLM development workflows.
Modelwire’s AI-generated summary of coverage from Simon Willison.
Modelwire analysis
Analyst takeOur AI-generated reading of the wider context and the next developments to watch.
The real signal isn't the CLI refresh itself, but that token counting is graduating from a developer convenience into a first-class monitoring primitive. The model-listing command and uvx integration suggest token accounting is becoming less of an afterthought and more of a deployment requirement.
This lands in the middle of a token-aware infrastructure moment. OpenAI's Decisions API (October 7) explicitly prices by token volume at a different tier, signaling that token consumption is now a product lever, not just a cost metric. Simultaneously, research from October 7 shows that reasoning models spike tokens when generating false outputs, making token counts a potential safety signal. ttok's modernization reflects that developers now need token visibility not just for budget tracking but for model behavior auditing. The tool sits between the infrastructure layer (where reasoning models consume variable tokens) and the application layer (where Decisions API charges by token volume).
If Simon Willison or other maintainers add token-spike detection or anomaly alerting to ttok or related llm plugins within the next two quarters, that confirms token accounting is moving from cost visibility into operational safety. If major LLM platforms follow OpenAI's lead and introduce task-specific pricing tiers tied to token consumption by Q2 2027, token-level tooling becomes mandatory rather than optional.
This interpretation is generated from the summary above and the archive coverage cited below. Our methodology · Report an error
Coverage behind this analysis
These archive entries ground the connection in our analysis. They are ordered by source publication date, with links to our coverage and the original sources.
·The Decoder
OpenAI splits API into specialized decision-making tier, cuts pricing tiers to three
OpenAI has introduced a specialized API tier designed for classification and structured decision-making tasks, processing queries roughly ten times faster than its general-purpose Responses API while charging a fraction of the cost at $0.10 per million tokens. The move signals a strategic shift toward task-specific inference optimization, allowing developers to offload binary choices, category selection,…
MentionsSimon Willison · ttok · OpenAI · tiktoken · uvx
How this coverage is produced
Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.
Modelwire summarizes, we don’t republish. Simon Willison originally reported this story as “ttok 0.4”. The full content lives on simonwillison.net. If you’re a publisher and want a different summarization policy for your work, see our takedown page.