Cheese Dossier · CF-1.3

Qwen3.6 Plus

qwen · Cheese #68· 6 signals · specialty gpqa · Proprietary

CHEESE INDEX
52.7
DEPTH
64.3
BALANCE
81.4
CONSENSUS
80.0
STABILITY
91.9
COVERAGE
92%
VALUE
15.06
I / O
$0.5 / $3
SPECIALTY
gpqa +15.4
BEST FIT
long_horizon 63.2

深度雷達

codingagentmathknowleinstrumultimmultilreason

實線=本模型 · 虛線=群體中位

芝士軸

coding40.4
agent52.5
horizon59.0

Cheese Index = 可用軸中位數;Balance = 三軸分歧小。

能力細節

swe ★78.8
gpqa ★90.4
hle28.8
mcp74.1
coding52.6
agent52.5
math62.4
knowledge57.9
instruction ★85.5
multimodal66.3
multilingual ★69.7
reasoning60.1

部署建議

中段指數:適合成本敏感路徑。優先看 Pareto/Value;用深度雷達找專長匹配,再與更高指數模型 A/B。

鄰近差分

vs Kimi K2.6

coding-2.0
agent+5.8
horizon+6.4