Unverified50% confidenceBenchmarkExact time
Under the Chat configuration (single session), at 262K context the prompt processing is 800-970 t/s and generation is 30-40 t/s; at 131K context the prompt processing is 930-1,040 t/s and generation is 30-45 t/s.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/26/2026
First Seen
Valid until: 11/9/2026
Sources
Related Entities
Related Claims
UnverifiedAccording to hands-on information shared by a Bilibili creator, on a hardware platform with 128G unified memory, 873G bandwidth, and 2070 TOPS of compute, a Qwen 30B-class model achieved over 100 TPS in a single stream on coding tasks.69% similarUnverifiedIn a pure CPU environment (2018 Threadripper, under load), SteerTTS edge achieves an RTF of 0.053 with 4 threads (about 19x real-time), with first-packet audio latency of about 120 ms thanks to sentence-level streaming.65% similarUnverifiedOn M4 Max, the text route using Omni 3B + TTS 1.7B (4.61 GB) has a median full-turn latency of 4.27 seconds, with a peak process RSS of 3.86 GiB.65% similarUnverifiedA reproducer measured HyperQwen's real chat-scenario throughput on an RTX 3090 at about 118 tok/s (greedy decoding averaged 118.1 tok/s across four prompts, range 111-131; 123.3 tok/s under temperature 0.7 sampling), far below 381 tok/s.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1099107API
curl https://kongchang.com/api/v1/knowledge/claims/1099107MCP
get_claim(id=1099107)