# batchwatch - full text > Measured queue times for the LLM batch APIs at OpenAI, Anthropic and > Google. Those endpoints cost 50% of the synchronous ones, but the only > promise is "within 24 hours". We time real jobs and publish the > distribution behind that ceiling. This is the full-text companion to https://batchwatch.dev/llms.txt. The short file is an index; this one carries the whole measured table, the method, and the API contract in one place, so a model that reads one file instead of crawling has everything without fetching a page. Generated: 2026-08-30T22:25:05.000Z (UTC). Rolling 30-day window; numbers recomputed every 10 minutes. Measurements span 2026-08-14T11:51:37.000Z to 2026-08-30T21:43:31.000Z. ## Method - how a number here is made Every figure is measured, never modelled. A duration is the difference between two server-side timestamps that batchwatch itself writes: one when a batch job is opened, one when the provider reports it finished. A client cannot submit a duration - only open and close a real job - so a fabricated eight-hour wait would cost eight real hours. What is excluded, and never counted as a completion: jobs that were abandoned, jobs still running, and jobs the caller cancelled. A percentile is computed only over completed jobs. The numbers are the public tier: a fixed 900-second delay, the same a keyless caller sees on /v1/curve. Paying customers get the live number; these are deliberately a step behind. Each figure carries n (how many completed jobs it rests on) and the cadence behind it - measured continuously in a rolling 30-day window. The full provenance of every figure, including its source count and a graded confidence, is live in /v1/coverage. Where we have measured nothing, we say so rather than estimate. ## Measured table (30-day window) ### anthropic/claude-haiku-4-5 - Provider: Anthropic - Median (p50): 2 min - Nine in ten (p90): 3 min - Slowest one in twenty (p95): 4 min - Longest measured: 9 min - Based on: 534 completed jobs, measured continuously since 2026-08-25 - Precision: 534 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-30T21:42:25.000Z - Oldest measurement in window: 2026-08-25T13:31:43.000Z - Page: https://batchwatch.dev/m/anthropic/claude-haiku-4-5 ### google/gemini-3.7-flash - Provider: Google - Median (p50): 3 min - Nine in ten (p90): 10 min - Slowest one in twenty (p95): 12 min - Longest measured: 12 min - Based on: 448 completed jobs, measured continuously since 2026-08-25 - Precision: 448 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-30T21:43:31.000Z - Oldest measurement in window: 2026-08-25T13:52:34.000Z - Page: https://batchwatch.dev/m/google/gemini-3.7-flash ### google/gemini-2.5-flash - Provider: Google - Median (p50): 2 min - Nine in ten (p90): 5 min - Slowest one in twenty (p95): 7 min - Longest measured: 14 min - Based on: 238 completed jobs, measured continuously since 2026-08-14 - Precision: 238 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-21T02:39:28.000Z - Oldest measurement in window: 2026-08-14T11:51:37.000Z - Page: https://batchwatch.dev/m/google/gemini-2.5-flash ### google/gemini-2.5-flash-lite - Provider: Google - Median (p50): 81s - Nine in ten (p90): 3 min - Slowest one in twenty (p95): 4 min - Longest measured: 5 min - Based on: 53 completed jobs, measured continuously since 2026-08-16 - Precision: 53 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-20T09:20:10.000Z - Oldest measurement in window: 2026-08-16T04:30:51.000Z - Page: https://batchwatch.dev/m/google/gemini-2.5-flash-lite ### openai/gpt-5.6-luna - Provider: OpenAI - Median (p50): 5 min - Nine in ten (p90): 5 hours - Slowest one in twenty (p95): 6 hours - Longest measured: 11 hours - Based on: 564 completed jobs, measured continuously since 2026-08-25 - Precision: 564 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-30T21:41:32.000Z - Oldest measurement in window: 2026-08-25T08:54:17.000Z - Page: https://batchwatch.dev/m/openai/gpt-5.6-luna ### openai/gpt-5.6-sol - Provider: OpenAI - Median (p50): 79s - Nine in ten (p90): 10 min - Slowest one in twenty (p95): 28 min - Longest measured: 66 min - Based on: 523 completed jobs, measured continuously since 2026-08-25 - Precision: 523 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-30T21:42:34.000Z - Oldest measurement in window: 2026-08-25T13:23:56.000Z - Page: https://batchwatch.dev/m/openai/gpt-5.6-sol ### openai/gpt-5-nano - Provider: OpenAI - Median (p50): 87s - Nine in ten (p90): 3 hours - Slowest one in twenty (p95): 4 hours - Longest measured: 7 hours - Based on: 508 completed jobs, measured continuously since 2026-08-25 - Precision: 508 completed jobs, measured continuously in a rolling 30-day window (median and p90 above) - Newest measurement: 2026-08-30T21:41:55.000Z - Oldest measurement in window: 2026-08-25T13:39:43.000Z - Page: https://batchwatch.dev/m/openai/gpt-5-nano Live JSON, always current: https://batchwatch.dev/v1/coverage ## Licence and citation These measurements are open data under Creative Commons Attribution 4.0 (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/. You may quote, republish and build on any number here, including commercially and including in an AI answer, for the price of a credit and a link back. That is the whole deal - we want these numbers used, and the licence exists to make that use visible, not to restrict it. Please attribute as: Measured by batchwatch - https://batchwatch.dev Or in HTML: Measured by batchwatch Full terms and the free/paid boundary: https://batchwatch.dev/license The open, delayed aggregates above are the CC BY data. Live, undelayed access under a paid plan is governed by the Terms (https://batchwatch.dev/terms), not by CC BY. ## API contract The read API (open, keyless, delayed like the numbers above): GET https://batchwatch.dev/v1/coverage Every measured model: n, contributors, median, confidence. This file is generated from it - if a number here disagrees, this file is the stale one. GET https://batchwatch.dev/v1/wait?provider=openai&model=gpt-5.6-sol How long to wait for one model, now. GET https://batchwatch.dev/v1/should-i-batch?provider=openai&model=gpt-5.6-sol Batch or run synchronously, for one job. GET https://batchwatch.dev/v1/estimate-batchtime?provider=openai&model=gpt-5.6-sol A point estimate with its interval. The write API - two ways in, not equally trusted: POST https://batchwatch.dev/v1/calls Open a measurement when the job starts. PATCH https://batchwatch.dev/v1/calls/{id} Close it when it lands. The duration is the gap between our two timestamps. POST https://batchwatch.dev/v1/calls/complete Import a finished job on YOUR timestamps. Requires a key - without one there is no physics behind the number. A key is not needed to contribute, but it is needed to count: keyless measurements are pooled as one anonymous source in the robust percentile, and only a key can earn a vote. ## Pages - https://batchwatch.dev/m/anthropic/claude-haiku-4-5 - how long claude-haiku-4-5 batch takes at Anthropic - https://batchwatch.dev/m/google/gemini-3.7-flash - how long gemini-3.7-flash batch takes at Google - https://batchwatch.dev/m/google/gemini-2.5-flash - how long gemini-2.5-flash batch takes at Google - https://batchwatch.dev/m/google/gemini-2.5-flash-lite - how long gemini-2.5-flash-lite batch takes at Google - https://batchwatch.dev/m/openai/gpt-5.6-luna - how long gpt-5.6-luna batch takes at OpenAI - https://batchwatch.dev/m/openai/gpt-5.6-sol - how long gpt-5.6-sol batch takes at OpenAI - https://batchwatch.dev/m/openai/gpt-5-nano - how long gpt-5-nano batch takes at OpenAI - https://batchwatch.dev/docs - API reference - https://batchwatch.dev/docs/query - the read API in full - https://batchwatch.dev/docs/ingest - how to contribute a measurement