{"slug": "anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb", "title": "Anthropic quadruples Claude API request cap to 128 MB from 32 MB", "summary": "Anthropic raised the maximum request size for the Claude Messages API and Token Counting endpoint from 32 MB to 128 MB, a fourfold increase the company announced in a post on X. Anthropic framed the change as a fix for developers whose image- and PDF-heavy requests hit 413 errors while context window remained, since base64 encoding inflates binary attachments by roughly a third and a 24 MB PDF already approached the old 32 MB ceiling. Anthropic's documentation as of this week still listed 32 MB for Messages and Token Counting, alongside 256 MB for the Message Batches API and 500 MB for the Files API, and the change does not affect pricing or the context window itself.", "body_md": "*The old ceiling was 32 MB, and Claude's API requests kept bumping into it. Anthropic has now quadrupled it to 128 MB.*\n\nPicture a long computer-use session, screenshot after screenshot piling into one request, or a stack of large PDFs sent in a single call. Under the old 32 MB limit, those requests failed while there was \"plenty of context window left\" to spare. Anthropic said this week that it raised the maximum request size for the Messages API and the Token Counting endpoint on the Claude API from 32 MB to 128 MB, a fourfold jump. The company announced the change in a post on X, framing it as a fix for a specific, recurring complaint from developers running image-heavy requests.\n\nThat detail is the real story here, not the number itself. Anthropic's context window, the amount of text and images a model can actually reason over, has nothing to do with the 32 MB figure developers kept running into. The request size cap measures the raw bytes of the HTTP payload hitting the API, not tokens. A document-heavy agent session could be nowhere near its token budget and still get bounced with a 413 error simply because the file attachments pushed the request body past 32 megabytes once base64-encoded.\n\nBase64 encoding is the part of this that made the old limit bite so early. Encoding binary files like images and PDFs for transport inflates their size by roughly a third, so a 24 MB PDF was already flirting with the 32 MB ceiling before a single token of instructions or conversation history got added. For a multi-turn computer-use session, where Claude takes repeated screenshots of a screen to decide its next click, those screenshots stack up fast inside one request even though the actual reasoning load barely registers against a 200,000-token or 1-million-token context window.\n\nAnthropic's wording in the X post, plenty of context window left, is a tell. It means developers were paying for a large context window, a feature Anthropic has marketed heavily, and then losing access to it for a completely unrelated reason: the size of the wrapper around the data, not the data's conceptual weight. That's a frustrating place for a developer to get stuck, because the error looks like a usage problem when it's actually an infrastructure ceiling that has nothing to do with how much Claude can think about.\n\n[Claude Opus 5.5 helped verify what may be mathstory's biggest single day](https://startupfortune.com/claude-opus-55-helped-verify-what-may-be-mathstorys-biggest-single-day/)\n\nAnthropic says an internal Claude research model rederived the Exponential Time Hypothesis refutation on October 16, ten days before OpenAI's announcement, using roughly 14 million output tokens and no human steering. Reviewers David Gross and Arthur Jaffe said their involvement should not be taken as certification of the claims. - [Claude Opus 5.5 verified math breakthrough independently](https://startupfortune.com/claude-opus-55-helped-verify-what-may-be-mathstorys-biggest-single-day/) - [OpenAI Anthropic same day computer science proofs](https://startupfortune.com/claude-opus-55-helped-verify-what-may-be-mathstorys-biggest-single-day/)\n\nDeveloper complaints about exactly this gap have surfaced in Anthropic's own public issue trackers, where the 32 MB cap was described as hard-stopping entire conversations rather than letting the model recover or truncate gracefully. Anthropic's own documentation, as of this week, still lists 32 MB as the limit for Messages and Token Counting, alongside 256 MB for the Message Batches API and 500 MB for the Files API, so the jump to 128 MB targets precisely the tier that image-heavy, interactive workloads actually use.\n\nFour times the room means a computer-use agent can run roughly four times as many screenshot-laden turns, or a workflow can attach a meaningfully larger PDF, before running into the same wall again. It doesn't touch pricing, and Anthropic didn't announce any change to the context window itself in the same post. The fix is narrower than that, and more useful for it: it closes the gap between what the model could already process and what the transport layer would let through.\n\nFor teams building document analysis tools or autonomous computer-use agents, this is the kind of change that shows up as fewer retries and fewer workarounds, like chunking a PDF into separate requests just to dodge a limit that had nothing to do with comprehension. It's a small, specific fix. It's also an admission that the 32 MB number was never really about what Claude could handle, just about what could get to it.\n\n**Also read:** [Bloom Energy posted record profit and Wall Street sold the stock anyway](https://startupfortune.com/bloom-energy-posted-record-profit-and-wall-street-sold-the-stock-anyway/) • [Caterpillar posted record revenue and backlog but its stock still fell](https://startupfortune.com/caterpillar-posted-record-revenue-and-backlog-but-its-stock-still-fell/) • [How To Negotiate An AI Agent Benchmark Pricing Clause](https://startupfortune.com/how-to-negotiate-an-ai-agent-benchmark-pricing-clause/)\n\n*This article is posted in [AI News](https://startupfortune.com/category/ai/), check it out for more related stories.*\n\n## Join the discussion\n\n[Open in the community →](https://startupfortune.com/community/)\n\nAlmost there. Sign in and your reply posts straight away.", "url": "https://wpnews.pro/news/anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb", "canonical_source": "https://startupfortune.com/anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb/", "published_at": "2026-10-30 01:37:32+00:00", "updated_at": "2026-10-30 01:46:48.916733+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-tools"], "entities": ["Anthropic", "Claude", "Claude API", "Messages API", "Token Counting endpoint", "Message Batches API", "Files API"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb", "markdown": "https://wpnews.pro/news/anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb.md", "text": "https://wpnews.pro/news/anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb.txt", "jsonld": "https://wpnews.pro/news/anthropic-quadruples-claude-api-request-cap-to-128-mb-from-32-mb.jsonld"}}