Unverified50% confidenceBenchmarkExact time
The model's regular stable decoding speed on the test platform is approximately 241 tokens/second.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/8/2026
First Seen
Valid until: 10/22/2026(expired)
Sources
Related Entities
Related Claims
VerifiedIn decode speed tests, speeds of approximately 45 tokens/second are achievable within 20,000 tokens, dropping sharply to approximately 20 tokens/second beyond 30,000 tokens.74% similarUnverifiedIn this test, when the 7900XTX is run with a 256K context, a single generation takes up to 743 seconds.71% similarUnverifiedIn this test, the prefill speed was fixed at approximately 500–600 tokens/s, and the time to first token increased linearly with input length.71% similarUnverifiedIn this test, the model's output speed in short-context scenarios is approximately 20 tokens/second, dropping to approximately 13 tokens/second for contexts of 200K+.70% similarUnverifiedIn the tests, when the context length exceeded 30,000 tokens, the prefill speed curves of the two configurations gradually converged.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/987100API
curl https://kongchang.com/api/v1/knowledge/claims/987100MCP
get_claim(id=987100)