self-hosted/ai
§01·model · /models

gpt-oss 20B

llmactiveApache-2.0

20B open-weight LLM by OpenAI (gpt-oss) for local reasoning and agentic use. Apache-2.0.

§02·GPUs that run this model
23 total
GPUVRAMSeriesBest speedMin VRAMWorksBenchmarksRecipe
RTX 4080 Super16GB406364prefill tokens/s16GB1recipecheck ↗
RTX 509032GB50298.2tokens/s3recipecheck ↗
RTX 508016GB50172.4tokens/s16GB2recipecheck ↗
RTX 3090 Ti24GB30160.3tokens/s24GB2recipecheck ↗
RTX 5070 Ti16GB50156tokens/s16GB2recipecheck ↗
RTX 309024GB30147.5tokens/s24GB2recipecheck ↗
RTX 408016GB40136.5tokens/s16GB2recipecheck ↗
RTX 4070 Ti Super16GB40128.9tokens/s16GB2recipecheck ↗
RTX 5060 Ti16GB5092.1tokens/s16GB2recipecheck ↗
RTX 306012GB3064tokens/s1recipecheck ↗
RTX 4060 Ti 16GB16GB4063.2tokens/s16GB2recipecheck ↗
Apple M2 Max64GBapple~0recipecheck ↗
Apple M2 Pro16GBapple~0recipecheck ↗
Apple M3 Max48GBapple~0recipecheck ↗
Apple M4 Max48GBapple~0recipecheck ↗
RTX 3080 Ti12GB30~0recipecheck ↗
RTX 407012GB40~0recipecheck ↗
RTX 4070 Super12GB40~0recipecheck ↗
RTX 4070 Ti12GB40~0recipecheck ↗
RTX 409024GB40~0recipecheck ↗
RTX 507012GB50~0recipecheck ↗
RX 7800 XT16GBamd~0recipecheck ↗
RX 7900 XTX24GBamd~0recipecheck ↗

benchmarked·~ runs via recipe (not benchmarked)· untested·doesn't fit