NVIDIA·graphics card
GeForce RTX 5060 Ti
Not enough owner reviews for a score yet
We've found 8 credible owner reviews for the GeForce RTX 5060 Ti so far, and need at least 10 before a score means anything. New reviews are collected every day.
Owner reviews
Overall owner mood
Owners mostly like it, with reservations: 38% of 8 reviews are positive, 49% mixed, and only 13% negative.
What owners like
Nothing two or more owners agree on yet.
What owners complain about
Nothing two or more owners agree on yet.
Built from the reviews themselves — every quote links to the owner who wrote it.
Full text of every credible post we scored. Bot-like and off-topic comments are hidden.
[...] ---- Yes, it uses a lot of power, but my VEGA64 is about as fast as an RTX 5060Ti(8gb) for using Ollama, locally (eight year old design, but it used HBM2 !!). [...]
[...] Less model coverage, but good prices and quite fast. [...] No one will give you as many input tokens for so little money as Claude Max x5 (maybe x20 too, I use x5). [...] Even with cache were talking thousands of $ in API prices a week. Hundreds if we're talking cheap cloud like Deepinfra. [...] For example my system a ryzen 7950x 192GB ram, 5x rtx3090 plus an rtx5060 ti 16gb. [...] If I include the rtx5060 in the mix I can bump to 5 users, but it slows down by about 15% (note the speeds are give are for one active user, multiple users at once see maybe 70% of tgat per user so aggregate is much higher in multi user setup). [...]
On my desktop RTX 5060 TI (16GB) and 96GB ram, I routinely get 25-30 tokens/sec using an 80B model quantized to int8. [...]
I tried it with Qwen/Qwen3.5-27B UD-IQ3XXS on an RTX 5060 Ti with 16Gb - the convert step succeeded, but upon loading the converted model I encountered: Model load deferred: 'Qwen35Config' object has no attribute 'vocabsize' My transformers version is up to date, and using AutoModelForCausalLM.frompretrained() should not raise this exception - what's going on here?
I have a RTX 5060TI that works smooth as butter on Arch+X11. Mix of office tasks, gaming on 1440p and AI workloads all without issues.
[...] Runs OK nicely on a RTX5060Ti, though I would have liked to have a bit higher thoughput (a 5080/5090 would do). [...]
I've got an R9700 32GB and an RTX 5060 Ti 16GB plus 64GB of system ram. [...] Anything below that isn't usable as a daily driver since at deep context it drops quite significantly, so if you start out at say 20 t/s then you'll wind up at like 10 t/s and 20 t/s is already too slow.
I have 2080 Ti mainly for development and I'm now thinking of buying another PC with RTX 5060 Ti for running local LLMs. Comparing the 5060 Ti (16GB) vs 5070 Ti apart from half the price, the former GPU only has half memory bus width of 128 instead of 256 bit for the latter. For running local LLMs is it really worth paying double the price of the 5070 Ti?