Expired50% confidenceBenchmarkExact time
In a local inference speed comparison, Agent A1 (95 tokens/second) was faster than Gemma 4 Coder and the Qwen series.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/22/2026
First Seen
Valid until: 10/20/2026(expired)
Sources
Related Claims
UnverifiedIn the BugFind code bug-finding test, Qwen3.6 27B Q4 precision scored 87 points, higher than Q6's 80 points, exhibiting a quantization reversal phenomenon.67% similarUnverifiedOn the HLE science benchmark, Agent A1 surpassed Qwen 3.6 and Step 3.5 Flash according to actual test data.67% similarUnverifiedQwen 3.6 35B A3B has a clear advantage over Gemma 4 and Qwen 3.5 in long-context memory benchmarks.66% similarUnverifiedIn an offline test across three phones, the iQOO 15 ran the Gemma 4 1/2B model at the fastest speed.64% similarUnverifiedIn actual testing using the Qwen3 27B Q4 quantized dense model, the generation speed with thinking mode disabled is approximately 58–60 tokens/s.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/593892API
curl https://kongchang.com/api/v1/knowledge/claims/593892MCP
get_claim(id=593892)