Unverified50% confidenceBenchmarkExact time
Real-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).
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/23/2026
First Seen
Valid until: 12/22/2026
Sources
Related Entities
Related Claims
UnverifiedWhen running 4-bit DeepSeek R1 on the M3 Ultra, a 1,000-token context generates approximately 19 tokens per second using 392GB, dropping to approximately 15 tokens per second using 410GB at a 16,000-token context.77% similarUnverifiedIn the STM32 embedded development test, the locally deployed DS4 model had an average output speed of approximately 23 tokens/second, memory usage spiked to 110 GB during inference, and the peak speed reached 83.8 tokens/second.76% 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).71% similarUnverifiedUnder a 4×RTX 3090 SXM4 configuration, the quantized version of GLM-5.2 achieves a token generation speed of approximately 6.3–6.8 tokens/second and a prompt evaluation speed of approximately 10.6 tokens/second.71% similarUnverifiedWhen running Qwen3-Next-Flash (Autoround W4A16 quantization), the prefill speed was approximately 1.3k tokens/s, and the generation speed reached 70 tokens/s (code) / 60 tokens/s (prose), with support for 128k+ long context.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/944878API
curl https://kongchang.com/api/v1/knowledge/claims/944878MCP
get_claim(id=944878)