Unverified50% confidenceBenchmarkExact time
In a failure case using Qwen3 0.6B as the drafter for a 4B model with the speculative token count set to 4, throughput crashed from 58.2 tokens/s to 16 tokens/s, and from 700 to 240 at 16 concurrency.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedIn the scenario where 16 requests each carry a 3,072-token long prompt, Gemma 4 E2B maintains 1049 tokens/second, while Granite 3.3 2B drops to 732 (0.70x), SmolLM3-3B drops to 725 (0.69x), Phi-4-mini drops to 404 (0.39x), and Qwen3-4B reaches only 373 (0.36x).78% similarUnverifiedIn the 16-concurrency, 256-token-output test, Gemma 4 E2B achieves an output speed of 1887 tokens/second, and the models faster than it include Qwen2.5-0.5B (4176, 2.21x), Qwen3-0.6B (3210, 1.70x), Qwen2.5-1.5B (2780, 1.47x), and Qwen3-1.7B (2049, 1.09x).74% similarUnverifiedOn a single H100 and at similar measured accuracy, QwenDLM-4B + ORL reaches 1,298 tokens/s, DFlash 746 tokens/s, and EAGLE3 405 tokens/s (end-to-end system comparison: ORL runs on dlmrt, speculative decoding runs on vLLM).74% similarUnverifiedIn Atomic Chat's Snake game test, Gemma 4 26b ran at approximately 136 token/s, while the dense Qwen3.6 27b ran at approximately 47 token/s.70% similarUnverifiedIn an ultra-long context test, Ornith 1.5 ran at around 75 tokens/second early on and dropped to approximately 45 tokens/second after exceeding 1.6 million tokens, while Qwen 3.8 dropped to approximately 5 tokens/second under the same load.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1071383API
curl https://kongchang.com/api/v1/knowledge/claims/1071383MCP
get_claim(id=1071383)