An experimental evaluation was conducted on the bench-scale test methods most commonly applied for generating data for fire toxicity assessments. The test methods evaluated were ISO/TS 19700, ISO 5660-1 with the controlled atmosphere box, and ISO 5659-2. Toxic gases were quantitatively analyzed using Fourier transfer infrared spectroscopy. Tests were made with 11 different insulation materials and polymethylmethacrylate as a reference material. The evaluation made was on the combustion conditions in the test apparatus, not generally on the precise yields measured. The evaluation focused on the ventilation conditions created in flaming combustion tests. It was seen that ISO/TS 19700 currently offers the best means among the three test methods evaluated for conducting tests at pre-determined and controlled equivalence ratios. The controlled-atmosphere cone calorimeter does not give a prolonged steady flaming combustion period for most materials and the influence of vitiation was difficult to predict and limiting in achieving higher equivalence ratios, with the test settings applied. ISO 5659-2 generally accumulates a mixture of gases from periods of both flaming and nonflaming combustion in a test, and the yields measured do not, in those cases, represent any specific combustion stage. For materials not showing flaming combustion, for example, mineral fiber products, the influence on the test conditions regarding oxygen consumption and heat generation from the material itself is limited compared to burning materials. However, there were specific properties and limitations of the different test methods observed that are important to consider.