Skip to main content

Can the RTX 4070 Ti SUPER Run gpt-oss-20b?

Runs

Yes. gpt-oss-20b at MXFP4 is a 12.1 GB download (Hugging Face file listing for ggml-org/gpt-oss-20b-GGUF), and the RTX 4070 Ti SUPER has 16 GB of VRAM, leaving about 3.9 GB for the KV cache and runtime. Published runs put generation at a median of 98.7 tokens per second across 5 cited measurements. Fastest cited run: Hardware Corner.

Model file12.1 GBMXFP4
VRAM16 GBRTX 4070 Ti SUPER
Headroom3.9 GBafter weights
Median speed98.7 tok/s5 cited runs
Bandwidth ceiling~323 tok/s672 GB/s ÷ 2.1 GB

Published runs for gpt-oss-20b on the RTX 4070 Ti SUPER

GenerationQuantRuntimeContextSourceDate
129 tok/s MXFP4 llama.cpp 4K Hardware Cornerlabelled "gpt-oss:20b" 2025-12-09
114 tok/s Q4_K_XL llama.cpp 16K Hardware Cornerlabelled "gpt-oss:20b" 2025-12-09
98.7 tok/s Q4_K_XL llama.cpp 32K Hardware Cornerlabelled "gpt-oss:20b" 2025-12-09
78.7 tok/s Q4_K_XL llama.cpp 64K Hardware Cornerlabelled "gpt-oss:20b" 2025-12-09
57.5 tok/s q4_K_XL llama.cpp 128K Hardware Cornerlabelled "gpt-oss:20b" 2025-12-09

More about this pair

How much VRAM does gpt-oss-20b need?

The MXFP4 file is 12.1 GB. Add roughly 1.5 GB for the runtime and an ~8K-token context, so about 13.6 GB of VRAM holds it comfortably. Longer contexts grow the KV cache and need more.

What is the fastest gpt-oss-20b can generate on the RTX 4070 Ti SUPER?

gpt-oss-20b is a mixture-of-experts model that reads about 2.1 GB of active weights per token. At the RTX 4070 Ti SUPER's 672 GB/s of memory bandwidth, that sets a ceiling near 323 tokens per second. Real runs land below it.

Check another pair

See the full matrix → · All RTX 4070 Ti SUPER benchmarks →

gpt-oss-20b on other GPUs

Other models on the RTX 4070 Ti SUPER

How the checks work

The fit is the MXFP4 file size against the card's VRAM, with 1.5 GB kept for the runtime and an ~8K context. A published run counts only if the pair fits, the run is Q4-class, its source does not call it an estimate, it is under the bandwidth ceiling above, and it is not one figure pasted across many cards. The whole matrix is downloadable as a CSV under CC BY 4.0.

As an Amazon Associate, SpecPicks earns from qualifying purchases.

This page is editorial synthesis based on publicly available information. No independent first-party benchmarking is reported.