TokenAssemble

OpenAI · open-weights model

gpt-oss-120b

gpt-oss-120b is a 117B-parameter model (MoE, 5.1B active) by OpenAI. Its smallest quant (MXFP4) needs at least 64.4GB of GPU-addressable memory at 8K context; it loads cleanly at a quality quant on 6 of the 50 machines we track.

Parameters
117B total · 5.1B active
Layers
36
Max context
128K
Attention
HYBRID
License
apache-2.0

Config from openai/gpt-oss-120b

Quants and what they need

Measured file sizes; “needs” = weights + KV cache at 8K context + overhead.

QuantFile sizeNeeds ≥
MXFP463.4 GB64.4 GB

What runs gpt-oss-120b

Every machine we track, graded at its best clean-fit quant.

HardwareBest-quant verdict · llama.cpp · 8K context
Intel Arc A770 16 GBTight · C · MXFP4interactive · offloadbeta
Intel Arc B580Tight · C · MXFP4interactive · offloadbeta
AMD Radeon PRO W7900Tight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 3060 12 GBTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 3070Won't fitbeta
NVIDIA GeForce RTX 3070 TiWon't fitbeta
NVIDIA GeForce RTX 3080 10 GBWon't fitbeta
NVIDIA GeForce RTX 3080 12 GBTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 3080 TiTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 3090Tight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 3090 TiTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4060Won't fitbeta
NVIDIA GeForce RTX 4060 Ti 16 GBTight · C · MXFP4usable · offloadbeta
NVIDIA GeForce RTX 4060 Ti 8 GBWon't fitbeta
NVIDIA GeForce RTX 4070Tight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4070 SUPERTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4070 TiTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4070 Ti SUPERTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4080Tight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4080 SUPERTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 4090Tight · C · MXFP4interactive · offloadbeta
NVIDIA RTX 5000 Ada GenerationTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 5060 Ti 16 GBTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 5070Tight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 5070 TiTight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 5080Tight · C · MXFP4interactive · offloadbeta
NVIDIA GeForce RTX 5090Tight · C · MXFP4interactive · offloadbeta
NVIDIA RTX 6000 Ada GenerationTight · C · MXFP4interactive · offloadbeta
NVIDIA RTX A6000Tight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 6800Tight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 6900 XTTight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 7800 XTTight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 7900 XTTight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 7900 XTXTight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 9070Tight · C · MXFP4interactive · offloadbeta
AMD Radeon RX 9070 XTTight · C · MXFP4interactive · offloadbeta
ASUS Ascent GX10Fits · A+ · MXFP4interactivebeta
Beelink GTR9 Pro Ryzen AI Max+ 395 128GBFits · A+ · MXFP4interactivebeta
NVIDIA DGX SparkFits · A+ · MXFP4interactivebeta
Framework Desktop Ryzen AI Max+ 395 128GBFits · A+ · MXFP4interactivebeta
Framework Desktop Ryzen AI Max+ 395 64GBWon't fitbeta
GMKtec EVO-X2 Ryzen AI Max+ 395 128GBFits · A+ · MXFP4interactivebeta
HP Z2 Mini G1a Ryzen AI Max+ PRO 395 128GBFits · A+ · MXFP4interactivebeta
Mac mini M4 16GBWon't fitbeta
Mac mini M4 24GBWon't fitbeta
Mac mini M4 Pro 24GBWon't fitbeta
Mac mini M4 Pro 48GBWon't fitbeta
Mac Studio M3 Ultra 96GBTight · C · MXFP4interactivebeta
Mac Studio M4 Max 36GBWon't fitbeta
Mac Studio M4 Max 64GBWon't fitbeta

Verdicts computed from published specs and measured GGUF sizes — how the math works.