Can RTX 5060 Ti 16GB run gpt-oss 20B?
Measured results for gpt-oss 20B (20.91B MoE, 3.6B active) on RTX 5060 Ti 16GB, plus what fits and the commands to run it.
Short answer
Yes. gpt-oss 20B Q8_0 fits fully on RTX 5060 Ti 16GB at 16k context, using 12.5 of 15.0 GB usable, and generates about 125.3 tokens/s in our own run. As the context fills it slows down: 111.3 tok/s with 16k tokens already in context (89% of the empty-context speed), measured. For Q8_0, our bandwidth-only estimate was ~118 tok/s against a measured median of ~125 (6% too pessimistic).
Measured generation speed by quantization
| Quant | Generation tok/s | Range | Prompt tok/s | Runs | Origin |
|---|---|---|---|---|---|
| Q4 class (median) | 134.1 | 134.1–134.1 | 5144 | 1 | includes our runs |
| Q8 class (median) | 125.3 | 125.3–125.3 | 5237 | 1 | includes our runs |
| Q4_K_M (our run, depth 0) | 134.1 | 3 repeats | 5144 | 3 | own measurement |
| Q8_0 (our run, depth 0) | 125.3 | 3 repeats | 5237 | 3 | own measurement |
Generation = tg128 (128 generated tokens, short context). Public rows are medians of community llama.cpp results (sources); our own runs use llama-bench b11658 on a rented GPU.
Speed vs context depth measured
Our own llama-bench runs on GeForce RTX 5060 Ti 16GB (flash attention on, 3 repeats): generation and prompt speed with the given number of tokens already in the context.
| Quant | Context depth | Generation tok/s | vs depth 0 | Prompt tok/s |
|---|---|---|---|---|
| Q4_K_M | 0 | 134.1 | 100% | 5144 |
| Q4_K_M | 4k | 125 | 93% | 4705 |
| Q4_K_M | 16k | 117.9 | 88% | 3674 |
| Q8_0 | 0 | 125.3 | 100% | 5237 |
| Q8_0 | 4k | 117.3 | 94% | 4747 |
| Q8_0 | 16k | 111.3 | 89% | 3680 |
Alternatives
Faster for gpt-oss 20B
- RTX 5080 16GB: Q8_0, 15 GB usable, ~239 tok/s measured
- RTX 5070 Ti 16GB: Q8_0, 15 GB usable, ~224 tok/s measured
- RTX 4090 24GB: Q8_0, 23 GB usable, ~258 tok/s measured
Other models on RTX 5060 Ti 16GB, similar size
- LFM2.5 24B-A2B (hardware page): Q4_K_M, fits fully, ~180 tok/s (scaled from measured)
- Gemma 4 12B: Q8_0, fits fully, ~31 tok/s (measured)
- Qwen3.5 9B: Q8_0, fits fully, ~46 tok/s (measured)
- Ornith 1.5 9B (hardware page): Q8_0, fits fully, ~46 tok/s (scaled from measured)
Memory and context limits
Q8_0 weights are 12.1 GB; the 16k f16 KV cache adds 0.41 GB. Largest context that fits in 15.0 GB usable:
| KV cache type | Max context |
|---|---|
| f16 | 64k |
| q8_0 | 128k |
| q4_0 | 128k |
Native limit 128k. Full KV cache table for gpt-oss 20B.
Copy-paste commands
The advisor's pick for this hardware: Q8_0, 16k context.
llama-server -hf unsloth/gpt-oss-20b-GGUF:Q8_0 -c 16384 -fa on --jinja -ngl 99
OLLAMA_CONTEXT_LENGTH=16384 ollama serve ollama run hf.co/unsloth/gpt-oss-20b-GGUF:Q8_0
FAQ
How fast is gpt-oss 20B on RTX 5060 Ti 16GB?
About 125.3 tokens/s for generation (Q8_0 on llama.cpp CUDA, 128-token generation), 5237 tokens/s for prompt processing. Speed drops as the context fills; see the depth table.
How much memory does gpt-oss 20B need on RTX 5060 Ti 16GB?
The advisor's pick, Q8_0, is 12.1 GB of weights plus 0.41 GB of KV cache at 16k context (12.5 GB), against 15.0 GB usable.
Is the gpt-oss 20B speed on RTX 5060 Ti 16GB measured or estimated?
Measured: our own llama-bench run on a rented GeForce RTX 5060 Ti 16GB, including runs with 4k to 16k tokens already in context.
What can I use instead of RTX 5060 Ti 16GB for gpt-oss 20B?
Nothing with less memory in our list fits it fully. Faster: RTX 5080 16GB (~239 tok/s), RTX 5070 Ti 16GB (~224 tok/s), RTX 4090 24GB (~258 tok/s).
What is the maximum context for gpt-oss 20B on RTX 5060 Ti 16GB?
With an f16 KV cache about 64k tokens, 128k with q8_0 and 128k with q4_0, bounded by the model's own limit.
Related
All models on RTX 5060 Ti 16GB · gpt-oss 20B on all hardware · VRAM tier · all measured pages · leaderboard
Compare: RTX 5060 Ti 16GB vs 16GB VRAM class (RTX 4060 Ti 16GB, 32GB RAM) · RTX 5060 Ti 16GB vs RTX 5070 Ti 16GB