Unverified50% confidenceBenchmarkExact time
During the prefill phase, Strata splits the prompt into chunks of up to 8,000 tokens and streams them over PCIe hidden behind computation, boosting the speed of processing a 32,000-token prompt from approximately 570 tokens/second to nearly 1,300 tokens/second.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/5/2026
First Seen
Valid until: 1/3/2027
Sources
Related Entities
Related Claims
UnverifiedWhen the context exceeds 64,000 tokens, Strata keeps most of the KV cache in memory and retains only the hottest 32,000 tokens in VRAM, improving speed from approximately 51 to 63 tokens/second at a 262,000-token context.80% 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% similarUnverifiedA 200K token context request requires approximately 25× the KV-Cache memory compared to an 8K token request.68% similarUnverifiedWhen 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.68% similarUnverifiedIf the quantized model weights are 150 GB, reading the weights once at a bandwidth of 273 GB/s takes approximately 0.55 seconds, giving a theoretical single-token generation speed of approximately 1.8 tokens/s (excluding computation latency and cross-node communication overhead).67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/978617API
curl https://kongchang.com/api/v1/knowledge/claims/978617MCP
get_claim(id=978617)