linear amplifier at X-band

Hi for every one actually I am designing a 9.4 GHz amplifier with 2 stages using PFB640, gain 18 dB with bandwidth 300 MHz but the problem I have the design unstable especially at the working frequency I added a resistors on the biasing circuit but it degraded the gain dramatically and I tried to optimized both keeping the gain with reasonable value and the stability factor little more than 1 but it not able to match the goals please I dont have too much time any one could help me and suggest a practical solutions because at the end I have to realize the board

First off we would need more insight into your design before we can make
and educated guess.

Simulated performance versus measured.
Circuit layout etc...

You are not using discrete components as matching elements, are you?

At 9GHz an easy way to see where the issue is, would be to get a piece of
copper foil (About the size of your RF traces) and move it around you circuit
till you see if you can kill the Osc / instability.

Also at 9GHz you better make sure your board manufacture has good enough
tolerances, or your whole design is unknown. A +/- 1 mil tolerance can be a killer
at high Frequencies.

I cant find a datasheet for that device so very difficult to help you without more info ... is it a single GaAsFET ? a GaAsFET MMIC (multistage) ?
do you have a link to a datasheet ? show us pix of board layoout and schematic you are using

cheers
Dave
VK2TDN

BFP640?