Unverified50% confidenceFactExact time
DeepSeek's generation speed is reportedly about 244 tokens per second, with early tests approaching 300 tokens per second.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/20/2026
First Seen
Valid until: 11/3/2026
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.79% similarUnverifiedThe model's regular stable decoding speed on the test platform is approximately 241 tokens/second.77% 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.75% similarUnverifiedUnder a citation prompt with a 29k-token context, HyperQwen decoding reached 237.6 tok/s and passed verbatim verification; prefill was about 1,210 tok/s, and the prompt completed in about 24 seconds, roughly five times that of llama.cpp (~250 tok/s).75% similarUnverifiedLong context has little effect on generation speed: after 4K context the median time per output token is 7.6 ms, and after 256K it is 8.3 ms; but the first read of a 128K-token prompt takes 18.33 seconds, so prefix caching is critical.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1077050API
curl https://kongchang.com/api/v1/knowledge/claims/1077050MCP
get_claim(id=1077050)