What servos to use will depend on the type of hexapod you want to build. You really need to study the hexapod kits that are available to see what servos are successfully used to build them.
It depends on the weight you need to lift, the road the leg has to travel, and the weight it needs to support while in position.
Without drawings, estimate on the weight and so on there is little you can advise.
My only advice is don't go for the cheapest servo's you can find.
From my previous experience in RC car racing, tower pro is probably the worst servo you can get. They are slow, unreliable and prone to burn out. (I've seen more than 1 burn out completely, which is pretty freaky when racing a 60kmp/h car filled up with nitromethanol).
The servo's are the thing that will make your hexapod more than a piece of plastic/metal/wood with an MCU on it. Don't go the cheapest way, pay a bit more and enjoy your pod all the more.
I don't have any experience in building a hexapod, but if I were to design one (or any small walking robot), I would go for servos with the max torque (kg-cm) I could afford, with metal gears, and most importantly metal bushings or full ball-bearings on both ends of the output shaft of the servos. Many cheap servos use no shaft support other than the plastic hole the shaft exits from as a bearing. Given the load placed on the servos, good shaft support is a must. Sintered bronze bushings (Oilite) would probably be ok for a small hexapod, but if full ball-bearings versions are available and only at a small amount to the cost of a single servo, spend the extra money, because it will be worth it in the long run. Finally, you only need to worry about servo speed if you are looking to have your robot scampering around like a crazed insect; you may find the speed of a servo not worth the extra money (or you might find youself spending money on something you don't end up using) - but that part is ultimately your call, as only you know what you will want the robot to do in the end, of course.