Unverified50% confidenceBenchmarkExact time
Haiku 5.5 can reach 100 to 200 TPS across different providers, and the creator's tests using Claude Code consistently showed approximately 180 TPS.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/9/2026
First Seen
Valid until: 1/7/2027
Sources
Related Entities
Related Claims
UnverifiedClaude Haiku 4.5拥有20万token的上下文窗口和6.4万token的最大输出长度70% similarUnverifiedOn a 128G compute module device, the decoding speed of Qwen3.8 Flash routinely reaches 241 TPS, with a peak of 274 TPS.62% similarVerifiedQwen2.5-Coder-32B running with Q4 quantization on an RTX 4090 can achieve an inference speed of 40–60 tokens per second.61% similarUnverifiedRunning a Qwen3 27B-level model on an AMD RX 7900XTX 24GB GPU, the output speed can reach up to 52–53 TPS, with an average stable speed of around 47 TPS.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/990912API
curl https://kongchang.com/api/v1/knowledge/claims/990912MCP
get_claim(id=990912)