@flue/runtime
2.3.1 · 18 Nov 2026Apache-2.0524.4k / weekRepositoryHomepagenpmAtom feed
2.3.1 is the newest release of 2.x.
In short
8 fixed
Release note
Patch Changes
Interrupted submissions no longer retry forever when the recovery render throws: once the submission's durability deadline passes, a submission whose classification render keeps failing is settled as timed out instead of being re-rendered on every supervisor wake without ever consuming an attempt.
The Cloudflare binding's Anthropic gateway path now maps the agent's
thinkingLevelto an adaptive-thinking effort (output_config.effort), souseModelthinking levels take effect on adaptive-thinking models instead of always running at Anthropic's default.The Cloudflare binding provider now accepts a
cacheRetentionoption ('short'or'long') to enable Anthropic prompt caching foranthropic/…gateway models — repeated prefixes are served from cache at the cached input rate instead of paying full input price every turn. Default'none'keeps the previous behavior.The Cloudflare Sandbox documentation page now presents
flue add sandbox cloudflareas a copyable prompt for your coding agent, with an explainer of what the blueprint-driven agent may do — installing@cloudflare/sandbox, wiring the Durable Object binding, migration, and containerDockerfile, and updating the agent to use the sandbox.Fix the first streamed delta being delayed by the full coalescing interval: after a quiet period, the first delta now flushes to the durable stream immediately, so observers see output as soon as the model starts responding.
Fix duplicate responses appearing after a model stream fails partway through and Flue retries it successfully. Clients now see only the successful replacement response instead of the incomplete first attempt followed by the complete retry.
Models synthesized from a dynamic model template (providers that serve model IDs beyond their catalog, such as Workers AI) are now detectable via the exported
isDynamicModel()helper, and the runtime warns once when such a model is first resolved — previously their cost silently read as $0 with no way to tell "free" apart from "unknown".GenAI trace content now stays schema-valid when it cannot be fully represented: oversized, unserializable, or transform-failing messages under
gen_ai.input.messages/gen_ai.output.messagesfall back to a shape-preservingrole: "flue"message (output fallbacks keepfinish_reason) instead of a bare diagnostic string or an array element that violates the message schema.
A whole project to check? Paste its package.json in the dependency check: every deprecated, outdated or vulnerable dependency, with what to upgrade to. Free, no account.