Unverified50% confidenceFactExact time
The Claude model family has historically been divided into a standard context of about 200K tokens and an extended long-context mode of up to 1 million tokens; long context is not enabled by default and usually requires extra billing or authorization under a specific plan.
1
Sources
50%
Confidence
Long-term
Relevance
10/26/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedClaude Code uses token-based consumption metering, with the Pro subscription having a fixed monthly cap, after which users must wait for the cycle to reset or pay extra.70% similarUnverifiedThe prefill throughput of this setup increases with prompt length, climbing from approximately 290 tokens/s for short prompts to approximately 1,300 tokens/s, with an average prefill throughput of approximately 900 tokens/s.63% similarUnverifiedMainstream model vendors allow the prefix portion to be cached, with cache-hit tokens priced at a substantial discount, but the cache has a lifetime and a minimum length requirement, and a hit only occurs when the prefix matches exactly.62% similarVerifiedOn most Claude models, Anthropic bills cache reads at 0.1× the base input price; default 5-minute cache writes at 1.25×; 1-hour cache writes at 2×; and a 5-minute cache entry is refreshed for free each time it is used.60% similarUnverifiedThe Claude API is billed at three tiers per million tokens—input, cached input and output—with no seat fees or monthly plans.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1101283API
curl https://kongchang.com/api/v1/knowledge/claims/1101283MCP
get_claim(id=1101283)