D
Best token generation
385,9 tok/s
DeepSeek-V4-Flash-284B-A13B
Best prefill
2.421,5 tok/s
DeepSeek-V4-Flash-284B-A13B
∅ Token generation
44,1 tok/s
Average of 38 runs
∅ Prefill
413,2 tok/s
Average of 38 runs
Best TTFT
1.320 ms
DeepSeek-V4-Flash-284B-A13B
Best harness rate
0%
no harness run yet
🏆
DeepSeek-V4-Flash-284B-A13B · fastest model at 385,9 tok/s token generation
Best prefill of this manufacturer: 2.421,5 tok/s (DeepSeek-V4-Flash-284B-A13B)
Best prefill of this manufacturer: 2.421,5 tok/s (DeepSeek-V4-Flash-284B-A13B)
Top models by token generation
Best measured generation throughput per model (tok/s), published performance benchmarks.