Unverified50% confidenceBenchmarkExact time
Naive-N0.5-Flash can be optimized to approximately 2000 tokens/s in single-concurrency scenarios, and can exceed 50 tokens per second in multi-concurrency scenarios
1
Sources
50%
Confidence
Long-term
Relevance
9/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUnder this configuration, the 35B model achieves a measured single-user peak speed of approximately 80 tokens/second, with concurrent throughput exceeding 500 tokens/second.73% similarUnverifiedThe UP host estimates that inference speed could improve to 30–40 tokens/s after using a GPU with 50GB of VRAM (not confirmed through actual testing).70% similarUnverifiedReal-world testing with the 284-billion-parameter DeepSeek V4 Flash shows token generation rising from 37 to 53 tokens/s (approximately 1.5x), and Prompt Processing jumping from 483 to 1,485 tokens/s (approximately three times).69% similarUnverifiedIn prefill speed tests, Halo achieved approximately 650 tokens/s while Spark achieved approximately 2000 tokens/s, with Spark being nearly three times faster.67% similarUnverifiedIn practical engineering, a chunk_size between 256 and 1024 tokens and a chunk_overlap between 20 and 100 tokens are fairly common empirical ranges.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/957116API
curl https://kongchang.com/api/v1/knowledge/claims/957116MCP
get_claim(id=957116)