gpt-oss 120B

96 records across 10 hardware.

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
H100gpt-oss 120B as-publishedvLLM119092––––2026-09-25community–source
H100gpt-oss 120B as-publishedvLLM102876––––2026-09-25community–source
H200gpt-oss 120B 120B MoEas-publishedvLLM275.84––––2026-03community–source
H200gpt-oss 120B 120B MoEas-publishedvLLM275.84––––2026-03community–source
H200gpt-oss 120B 120B MoEas-publishedvLLM231.22––––2026-03community–source
RTX Pro 6000 Blackwellgpt-oss 120B 120BMXFP4llama.cpp210.14578.6––4096–sourced–source
RTX Pro 6000 Blackwellgpt-oss 120B 120BMXFP4llama.cpp182.24060.7––16384–sourced–source
RTX Pro 6000 Blackwellgpt-oss 120B 120BMXFP4llama.cpp161.53368.3––32768–sourced–source
RTX Pro 6000 Blackwellgpt-oss 120B 120BMXFP4llama.cpp133.12360.3––65536–sourced–source
RTX Pro 6000 Blackwellgpt-oss 120B 120BMXFP4llama.cpp99.81289.8––131072–sourced–source
M5 Ultragpt-oss 120B MXFP4-Q8oMLX97.2858.1––10242026-10-08community–source
M5 Maxgpt-oss 120B MXFP4-Q8oMLX93.0885.3––10242026-10-15community–source
M5 Ultragpt-oss 120B MXFP4-Q8oMLX90.23148.0––81922026-10-08community–source
M5 Maxgpt-oss 120B MXFP4-Q8oMLX89.42061.0––40962026-10-15community–source
M5 Maxgpt-oss 120B MXFP4oMLX81.2727.7––10242026-10-11community–source
M5 Maxgpt-oss 120B MXFP4oMLX81.2727.7––10242026-10-11community–source
M5 Maxgpt-oss 120B MXFP4oMLX78.52110.0––40962026-10-11community–source
M5 Maxgpt-oss 120B MXFP4oMLX78.52110.0––40962026-10-11community–source
M5 Ultragpt-oss 120B MXFP4-Q8oMLX75.42941.0––327682026-10-08community–source
M3 Ultragpt-oss 120B 8bitoMLX74.3876.2––10242026-10-09community–source
M3 Ultragpt-oss 120B 8bitoMLX72.51072.0––40962026-10-09community–source
M3 Ultragpt-oss 120B 8bitoMLX69.61068.0––81922026-10-09community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX68.4613.2––10242026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX68.4613.2––10242026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX68.4613.2––10242026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX68.4613.2––10242026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX68.4613.2––10242026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX67.7314.1––10242026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX67.7314.1––10242026-03-08community–source
Arc Pro B70gpt-oss 120B MXFP4llama.cpp (SYCL)67.14––––2026-10-01community–source
M3 Ultragpt-oss 120B 8bitoMLX65.51011.0––163842026-10-09community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX64.1685.5––40962026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX64.1685.5––40962026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX64.1685.5––40962026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX64.1685.5––40962026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX64.1685.5––40962026-03-08community–source
M3 Ultragpt-oss 120B 8bitoMLX60.7569.3––10242026-03-08communitycontradictionsource
M3 Ultragpt-oss 120B 8bitoMLX60.7569.3––10242026-03-08communitycontradictionsource
M2 Ultragpt-oss 120B MXFP4-Q8oMLX60.1363.7––40962026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX60.1363.7––40962026-03-08community–source
M3 Ultragpt-oss 120B 8bitoMLX59.0915.4––327682026-10-09community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX57.8687.8––81922026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX57.8687.8––81922026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX57.8687.8––81922026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX57.8687.8––81922026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX57.8687.8––81922026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX54.9187.7––10242026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX54.9187.7––10242026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX54.9187.7––10242026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX54.9187.7––10242026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX54.9187.7––10242026-03-08community–source
M3 Ultragpt-oss 120B 8bitoMLX54.2631.9––40962026-03-08communitycontradictionsource
M3 Ultragpt-oss 120B 8bitoMLX54.2631.9––40962026-03-08communitycontradictionsource
M3 Ultragpt-oss 120B MXFP4-Q8oMLX47.5663.6––163842026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX47.5663.6––163842026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX47.5663.6––163842026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX47.5663.6––163842026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX47.5663.6––163842026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX44.7182.0––40962026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX44.7182.0––40962026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX44.7182.0––40962026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX44.7182.0––40962026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX44.7182.0––40962026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX43.7365.9––81922026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX43.7365.9––81922026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX38.4361.6––163842026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX38.4361.6––163842026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX37.1183.9––81922026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX37.1183.9––81922026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX37.1183.9––81922026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX37.1183.9––81922026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX37.1183.9––81922026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX35.2600.7––327682026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX35.2600.7––327682026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX35.2600.7––327682026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX35.2600.7––327682026-03-08community–source
M3 Ultragpt-oss 120B MXFP4-Q8oMLX35.2600.7––327682026-03-08community–source
M5 Maxgpt-oss 120B MXFP4oMLX34.11318.0––1310722026-10-11community–source
Arc Pro B70gpt-oss 120B MXFP4llama.cpp (SYCL)33.27––––2026-10-01community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX25.7337.1––327682026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX25.7337.1––327682026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX22.7177.1––163842026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX22.7177.1––163842026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX22.7177.1––163842026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX22.7177.1––163842026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX22.7177.1––163842026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX15.4294.5––655362026-03-08community–source
M2 Ultragpt-oss 120B MXFP4-Q8oMLX15.4294.5––655362026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX14.6165.3––327682026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX14.6165.3––327682026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX14.6165.3––327682026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX14.6165.3––327682026-03-08community–source
M2 Maxgpt-oss 120B 4bitoMLX14.6165.3––327682026-03-08community–source
M4 Ultra (192 GB)gpt-oss 120B Q4_K - MediumMLX13––4.2––sourced–source
Arc Pro B70gpt-oss 120B MXFP4llama.cpp (SYCL)–594.5–––2026-10-01community–source
Arc Pro B70gpt-oss 120B MXFP4llama.cpp (SYCL)–600.1–––2026-10-01community–source

← all models