Unverified50% confidenceFactExact time
170 tokens/s is the combined speed across all concurrent tasks, not the individual speed of a single agent.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/24/2026
First Seen
Valid until: 12/23/2026
Sources
Related Entities
Related Claims
VerifiedIn decode speed tests, speeds of approximately 45 tokens/second are achievable within 20,000 tokens, dropping sharply to approximately 20 tokens/second beyond 30,000 tokens.72% similarUnverifiedLocal model inference speed is approximately 1–10 tokens/second, while cloud-based APIs can reach 50–150 tokens/second.71% similarUnverifiedAfter enabling MTP, inference speed fluctuates significantly, sometimes exceeding 40 tokens/second and sometimes dropping to 25 tokens/second.64% similarUnverifiedUnder 16-way concurrent load, D-Flash2 achieves a total throughput of 227.6 tokens/s, an average of 28.2 tokens/s per channel, with a time-to-first-token latency of 4.18 seconds.63% similarUnverifiedAn agent task requiring 20 LLM calls with 200 tokens generated each time takes approximately 133 seconds to complete at 30 TPS, but only 13 seconds at 300 TPS.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/946669API
curl https://kongchang.com/api/v1/knowledge/claims/946669MCP
get_claim(id=946669)