Can I run gpt-oss-120b on the GeForce RTX 4070?
Won't runMXFP4 needs 63.6 GB, 52.2 GB more than the 11.4 GB of usable memory. With 64 GB of system RAM: Runs slowly, est. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%).
12 GB VRAM · 504.0 GB/s · FP16 29.1 TFLOPS. Assumes 32 GB of DDR5-5600 system RAM, Windows with this GPU driving the display, 8k context and an f16 KV cache. The GeForce RTX 4070 Super has the same memory setup (12 GB, 504.0 GB/s), so its generation-speed estimate is the same; only prompt processing differs.
Verdict by system RAM
This page assumes 32 GB of system RAM. For this pair, the verdict depends on how much system RAM you have.
- 16 GBWon't runMXFP4 doesn't fit
- 32 GBWon't runMXFP4 doesn't fit(used on this page)
- 64 GBRuns slowlyest. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%)(used for the commands below)
- 96 GBRuns slowlyest. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%)
- 128 GBRuns slowlyest. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%)
Every quant of gpt-oss-120b on the GeForce RTX 4070
| Quant | File size | Verdict | Speed | Memory | Runs as | Notes |
|---|---|---|---|---|---|---|
| MXFP4baseline | 63.39 GB | Won't run | — | Needs about 63.6 GB; 11.4 GB of VRAM and 28.0 GB of RAM are free | — | — |
Models with a thinking mode spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s). With thinking turned off, a model may do one grade better.
Where the memory goes at MXFP4
The memory, context and speed figures below assume 64 GB of system RAM (the configuration from the RAM hint above); with 32 GB nothing fits.
- Weights
- 8.5 GB
- KV cache
- 0.3 GB
- Compute buffer
- 0.6 GB
- OS reserve
- 0.6 GB
- Free
- 2.0 GB
- Weights in system RAM
- 54.9 GB
System RAM, context length and KV cache
| Context | KV f16 | KV q8_0 | KV q4_0 |
|---|---|---|---|
| 4k | 63.4 GBRuns slowly est. 22.3 tok/s15.6–29.0theoretical estimate ±30% | 63.4 GBRuns slowly est. 22.4 tok/s15.7–29.2theoretical estimate ±30% | 63.3 GBRuns slowly est. 22.5 tok/s15.7–29.2theoretical estimate ±30% |
| 8k | 63.6 GBRuns slowly est. 22.1 tok/s15.5–28.7theoretical estimate ±30% | 63.5 GBRuns slowly est. 22.3 tok/s15.6–29.0theoretical estimate ±30% | 63.4 GBRuns slowly est. 22.4 tok/s15.7–29.1theoretical estimate ±30% |
| 16k | 64.0 GBRuns slowly est. 21.2 tok/s14.9–27.6theoretical estimate ±30% | 63.7 GBRuns slowly est. 22.1 tok/s15.5–28.7theoretical estimate ±30% | 63.6 GBRuns slowly est. 22.3 tok/s15.6–29.0theoretical estimate ±30% |
| 32k | 64.8 GBRuns slowly est. 20.5 tok/s14.3–26.6theoretical estimate ±30% | 64.2 GBRuns slowly est. 21.2 tok/s14.8–27.5theoretical estimate ±30% | 63.9 GBRuns slowly est. 22.1 tok/s15.4–28.7theoretical estimate ±30% |
| 64k | 66.3 GBRuns slowly est. 18.8 tok/s13.1–24.4theoretical estimate ±30% | 65.2 GBRuns slowly est. 20.4 tok/s14.3–26.5theoretical estimate ±30% | 64.5 GBRuns slowly est. 21.1 tok/s14.8–27.5theoretical estimate ±30% |
| 128k | 69.3 GBdoes not fit | 67.0 GBRuns slowly est. 18.6 tok/s13.0–24.2theoretical estimate ±30% | 65.8 GBRuns slowly est. 19.9 tok/s13.9–25.8theoretical estimate ±30% |
S = Runs great · A = Runs well · B = Runs slowly · F = Won't run
With 64 GB of system RAM: At MXFP4 the verdict stays Runs slowly up to 64k context with an f16 KV cache, and up to 128k with q8_0.
Estimated speed
- Generation
- With 64 GB of system RAM: est. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%)
- Prompt processing
- Prompt-processing estimates only apply when the whole model runs on the GPU.
Why this verdict: Experts run from system RAM — Runs well needs 30 tok/s or more on this path
Measured on this exact combination
No public measurement for this exact combination yet; the numbers above are estimates.
If this is not enough
A smaller model that runs well on the GeForce RTX 4070
Runs wellQwen3.5 35B-A3B: Runs well, est. 57.1 tok/s (40.0–74.3, theoretical estimate ±30%)
Same GPU with 64 GB of system RAM
Runs slowlyMXFP4: Runs slowly, est. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%)
How to run it
Commands for the MXFP4 file. Both tools download from Hugging Face on first run.
These commands assume 64 GB of system RAM (the configuration from the RAM hint above). For a different amount of RAM, pick it under “System RAM, context length and KV cache” above to see the commands for it.
File: gpt-oss-120b-MXFP4.gguf (63.39 GB) from ggml-org/gpt-oss-120b-GGUF.
llama-server -hf ggml-org/gpt-oss-120b-GGUF:MXFP4 -c 8192 -ngl all --n-cpu-moe 32 -fa on- -ngl all loads every layer on the GPU.
- --n-cpu-moe 32 keeps the experts in the first 32 layers in system RAM; the remaining 4 layers with experts keep them on the GPU. This estimate assumes the same split.
- If loading fails with an out-of-memory error, raise the value to 33, or to 34 if that isn't enough.
- llama.cpp b7410 (December 2025) and later split the model in a similar way on their own if you leave out -ngl and --n-cpu-moe.
- -fa on enables flash attention (needed for KV cache quantization).
- --cache-type-k q8_0 --cache-type-v q8_0 shrinks the KV cache to 0.16 GB at 8k context.
PowerShell (quit the Ollama tray app first):
$env:OLLAMA_CONTEXT_LENGTH="8192"; ollama serve
ollama run hf.co/ggml-org/gpt-oss-120b-GGUF:MXFP4- Ollama defaults to 4k context on GPUs under 24 GB. Set OLLAMA_CONTEXT_LENGTH when starting the server, or /set parameter num_ctx inside the chat.
- Ollama 0.30 and later automatically keep part of the experts in system RAM to fit your VRAM. Older versions split whole layers instead, which is much slower.
/set parameter num_ctx 8192
Related pages
Guide: How much VRAM do local LLMs need? GPU requirements for 8–32 GB (2026)
Other models on the GeForce RTX 4070
gpt-oss-120b on other hardware
Frequently asked questions
Can I run gpt-oss-120b on the GeForce RTX 4070?
MXFP4 needs 63.6 GB, 52.2 GB more than the 11.4 GB of usable memory. With 64 GB of system RAM: Runs slowly, est. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%). MXFP4 needs 63.6 GB at 8k context; this setup has 11.4 GB of usable memory and 28.0 GB of free system RAM.
How much context can gpt-oss-120b use on the GeForce RTX 4070?
With 64 GB of system RAM: At MXFP4 the verdict stays Runs slowly up to 64k context with an f16 KV cache (2.4 GB of KV) and up to 128k with a q8_0 KV cache (2.6 GB). The model supports up to 128k.
Which quant should I use, and how fast is it?
No tracked quant runs with 32 GB of RAM; with 64 GB, MXFP4 reaches Runs slowly, est. 22.1 tok/s (15.5–28.7, theoretical estimate ±30%).
Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.