TL;DR
A developer rumor says Anthropic is close to releasing Claude Opus 4.8 and Sonnet 4.8, but the strongest cited market signal appears to have been misread. Thorsten Meyer AI says Polymarket showed about 26% odds for May 31 and higher odds for later dates, with no Anthropic announcement, model card, API ID or benchmark confirming Claude 4.8.
A rumor that Anthropic is about to release Claude Opus 4.8 and Sonnet 4.8 is spreading through developer circles, but the strongest cited evidence appears to be weaker than some posts suggest: Thorsten Meyer AI reports that Polymarket’s roughly 70% odds refer to a later June date, while the market price for a May 31 release was about 26% as of May 27, 2026.
The confirmed picture is narrower than the rumor. Anthropic has released Claude Opus 4.7, which Thorsten Meyer AI lists as having launched on April 16, 2026, priced at $5 per million input tokens and $25 per million output tokens, with a 1 million-token context window. Claude Sonnet 4.6 is listed as the current Sonnet model, launched February 17, 2026, at $3 per million input tokens and $15 per million output tokens, also with a 1 million-token context window.
The main public clue cited for a possible Claude 4.8 release is a sonnet-4-8 string that appeared in a March 31 Claude Code source map, inside what Thorsten Meyer AI described as a filter list for forbidden model names. That is evidence that the name appeared in developer-facing code, but it does not confirm a release plan, model capability, launch date or final product name.
Thorsten Meyer AI said there is no Anthropic announcement, API model identifier, model card or benchmark set confirming Claude 4.8. The outlet also described the Polymarket contract as a date ladder, not a single yes-or-no claim, with separate implied probabilities for several release deadlines. On that reading, the market was pricing a release as possible in the near term, but not strongly expected by the end of May.
Why It Matters
The distinction matters because developers and companies are making near-term choices about coding agents, model routing, cost controls and product roadmaps. If teams treat an unconfirmed Claude 4.8 release as imminent, they may delay deployments, evaluations or budget decisions around models that already exist.
The report also shows the risk of treating thin prediction markets as roadmap signals. Thorsten Meyer AI said the Polymarket contract had about $900 in total volume as of May 27, 2026. That makes it a useful gauge of trader sentiment, but a limited basis for claiming knowledge of Anthropic’s release schedule.
Choosing the Right AI Engine for Your Organization
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
The current rumor sits against Anthropic’s recent release pattern. Sonnet 4.6 launched on February 17, 2026, and Opus 4.7 followed on April 16, 2026. Thorsten Meyer AI notes that Anthropic has paired Opus and Sonnet minor versions in prior releases, but there has been no Sonnet 4.7 release. A direct move from Sonnet 4.6 to Sonnet 4.8 would break that pattern, unless Anthropic chooses to realign the Opus and Sonnet version numbers in a later release.
The bullish interpretation is that the leaked string, a steady cadence and rising market probabilities for later dates point toward a pending update. The bearish interpretation is that the evidence is still thin: no official signal, no delivered model and no public confirmation that sonnet-4-8 will ship under that name.
“The 70% number is real. But the two have been welded together wrong.”
— Thorsten Meyer AI
“May 31 is actually ~26%.”
— Thorsten Meyer AI
“No Anthropic announcement, no API id, no model card, no benchmarks for 4.8.”
— Thorsten Meyer AI
“A sentiment gauge, not a leaked roadmap.”
— Thorsten Meyer AI
developer AI model API access
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
It is not clear whether Anthropic will release Claude Opus 4.8, Claude Sonnet 4.8, both models, or neither under those names. It is also unclear whether the March sonnet-4-8 string reflected an active product plan, an internal placeholder, a blocked name pattern or another engineering artifact. No public benchmarks or capability claims for Claude 4.8 have been confirmed by Anthropic.
BXQINLENX Professional 8 PCS Model Tools Kit Modeler Basic Tools Craft Set Hobby Building Tools Kit for Gundam Car Model Building Repairing and Fixing(A)
● FUNCTION—EASY TO USE—The modeler basic tools set is suitable for a beginner and advanced modeler as well.You…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
The next concrete marker would be an Anthropic announcement, API model ID, pricing page update, model card or benchmark release. Until then, the practical guidance from the report is to evaluate existing Claude models: use Sonnet 4.6 for routine coding-agent work and Opus 4.7 for harder reasoning, refactoring and long-context tasks where the higher cost is justified.
Agentic AI Unleashed: A guide to designing, building, and deploying autonomous AI systems (English Edition)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Did Anthropic announce Claude 4.8?
No. Based on the source material, there has been no Anthropic announcement, API ID, model card or benchmark confirming Claude 4.8.
What was the source of the Claude 4.8 rumor?
The main cited clue is a sonnet-4-8 string found in a March 31 Claude Code source map, plus prediction-market pricing around possible release dates.
Does Polymarket show 70% odds for a May 31 release?
Thorsten Meyer AI says no. Its reading is that May 31 was priced around 26%, while the roughly 70% figure was tied to June 15.
Why is the missing Sonnet 4.7 release relevant?
Thorsten Meyer AI says Anthropic has paired Opus and Sonnet minor versions before. A direct Sonnet 4.6 to Sonnet 4.8 move would break that pattern, which makes the rumored naming less certain.
What should developers do while waiting?
The report suggests using models that are available now: Sonnet 4.6 for balanced coding-agent work and Opus 4.7 for harder tasks that need stronger reasoning or larger refactors.
Source: Thorsten Meyer AI
