gpt-oss 120B › NVIDIA RTX 4060
gpt-oss 120B on the NVIDIA RTX 4060
Best measured: 18.9 tokens/s generation and 333 tokens/s prompt processing (MXFP4, CUDA, FA on).
For comparison, our size-scaling estimate for this combination is 34.1 tok/s.
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| MXFP4 | CUDA FA on | 18.9 | 333 | tg128 | Djip007 llama-bench model column: gpt-oss 120B MXFP4 MoE; model file is larger than this card's VRAM, so part of it ran on the CPU/system RAM | llama.cpp discussion #15396 |
| F16 | CUDA FA on | 17.4 | 201 | tg128 | Djip007 llama-bench model column: gpt-oss 120B F16; model file is larger than this card's VRAM, so part of it ran on the CPU/system RAM | llama.cpp discussion #15396 |
tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128.
Doesn't fit in this card's VRAM
Part of this model ran on the CPU here. A rented 48-80 GB GPU keeps it all in VRAM.
Rent a bigger GPU on RunPodAffiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.
All results for the NVIDIA RTX 4060 · gpt-oss 120B on other hardware