RTX 5070
An RTX 5070 runs 11 of the 34 open models tracked here at Q4_K_M with 8k context, computed on 16 October 2026. The largest is Qwen3 14B, needing 10.8 GB and generating about 55 tokens per second.
- Memory
- 12 GB VRAM
- Bandwidth
- 672 GB/s
- Board power
- 250 W
- Released
- 2025
- Availability
- Sold new
Open the RTX 5070 in the LLM checker
What an RTX 5070 runs, model by model
RTX 5070 12 GB addresses 12 GB at 672 GB/s. Every model below is at Q4_K_M with 8k of context, the setting most people actually use. Once the 5% safety margin comes off, the card has 11.4 GB to spend: memory decides what loads, bandwidth how fast it answers. Both are estimates from stated formulas, and the VRAM calculator shows the working.
On this card a model fits at or under 11.4 GB, and above 10.2 GB it is tight: it loads, but there is little left for the conversation, and a desktop on the same 12 GB can push it over. Tokens per second are single-stream estimates from memory bandwidth; ≤ marks a ceiling real runtimes fall well below.
Computed 16 October 2026The model to actually run on it
Gemma 4 12B is the best use of this card: 8.2 GB of the 12 GB available, an estimated 68 tokens per second, faster than you can read, and 131,072 tokens of context still available.
How this is chosen, since no benchmark is quoted: models are compared by size, a mixture-of-experts model counting at the geometric mean of its total and active parameters (a rule of thumb, not a measurement). The pick is the biggest class that fits with memory to spare and answers at 25 tokens per second or better, and within that class the most recent general-purpose release; coding and reasoning specialists are listed above but not recommended by default.
Qwen3 14B is the largest model the card will hold, at 10.8 GB, but it is the wrong daily driver: filling 90% of memory leaves room for only 11,264 tokens of conversation, and it generates at 55 tokens per second against 68. More parameters are not worth a context window that runs out mid-document.
If quality matters more than parameter count, Qwen3 8B fits at Q8_0 in about 10.7 GB. Q8 is near-lossless; a smaller model at Q8 often beats a larger one squeezed into Q3.
How much context actually fits
Model size is the question people ask; context is the one that bites. The KV cache grows linearly with the conversation, so a model that loads comfortably can still run out of memory halfway through a long document. On 12 GB, the context you get is whatever the weights leave behind, which is why two cards that run the same model can feel completely different in use. These are the largest contexts this card holds, at Q4_K_M weights.
| Model | Max context, FP16 KV | With Q8 KV cache |
|---|---|---|
| Gemma 4 12B | 131,072 tokens | 131,072 tokens |
| Qwen3 14B | 11,264 tokens | 23,552 tokens |
| Phi-4 14B | 9,216 tokens | 16,384 tokens · model limit |
| Ministral 3 14B | 14,336 tokens | 29,696 tokens |
Quantising the KV cache to Q8 roughly doubles what you can hold, at a quality cost most people never notice: Qwen3 14B on this card goes from 11,264 to 23,552 tokens. Capped at 128k tokens, or at the model's own native window where that is smaller, marked "model limit": free memory beyond that point buys nothing. Rows that hold far more context than their size suggests are hybrid, sliding-window or latent-attention models, which cache only a fraction of what a standard transformer does; each model page shows the working.
Where this card stops
The first model out of reach is gpt-oss 20B: about 13.4 GB at Q4_K_M and 8k context, against 12 GB of usable memory. Dropping to Q3_K_M would need about 11 GB, which fits, though Q3 loses enough quality that a smaller model at Q4 is usually the better trade. Offloading the remainder to system RAM works and is ten to fifty times slower; it is a way to see a model run, not a way to use one.
Buy or rent
Worth it if your models fit in 12 GB and you also want the card for games. If models are the reason, a 16 GB card holds more, and a used 24 GB card holds far more. For the occasional larger model, renting by the hour is cheaper than buying up a tier.
Retail prices drift, and a figure written today would be wrong within a month, so this page does not carry one. Put the figure you are actually looking at, with your electricity rate and your real monthly hours, into the build vs rent calculator, which already knows this card draws 250 W under load. It returns the month owning becomes cheaper, or tells you it never does. Hours per month decides it far more often than the price does.
The honest take
Fast memory, and too little of it: 12 GB at 672 GB/s.
The 5070 reads memory a third faster than the 4070 it replaces and half again as fast as a 5060 Ti 16 GB, so whatever fits runs quickly. The catch is 12 GB, the same ceiling as a 3060 or a 4070, which stops it short of what a 16 GB card holds. It is a Blackwell card, so FP4 formats run natively and old software builds will not use it.
What it is good at
- 672 GB/s: a third faster than the 4070's 504 GB/s, and level with the RTX PRO 4000.
- A two-slot Founders Edition at 250 W, which suits most cases and power supplies.
- Blackwell generation, so FP4 weight formats run natively.
What it is not
- Twelve gigabytes: it holds what a 3060 or a 4070 holds, and any 16 GB card holds more.
- Blackwell needs CUDA 12.8 or newer builds of PyTorch and llama.cpp; an old install will not use the card.
- The extra speed cannot be spent on larger models, which is where most local AI upgrades go.
The thing people get wrong: The 5060 Ti 16 GB holds 4 GB more than the 5070. The 5070 reads memory half again as fast, 672 GB/s against 448 GB/s, but for language models capacity decides what runs at all, so the slower card runs models the faster one cannot hold.
Compared with the alternatives
- RTX 5060 Ti: 16 GB · 448 GB/s · best fit Gemma 4 12B at ~45 tok/s. The newest 16 GB GeForce below the high end, and a clear step up from the 4060 Ti it replaces.
- RTX 4070: 12 GB · 504 GB/s · best fit Gemma 4 12B at ~51 tok/s. Fast for its class, and capped by 12 GB.
- RTX 5070 Ti: 16 GB · 896 GB/s · best fit Gemma 4 12B at ~90 tok/s. The 5080's memory bus and 16 GB, read about 7% more slowly.
All 39 cards are compared side by side on the GPU index, and the same numbers from the model's point of view are on each model fit table. To check one specific pairing rather than read a table, use the can-I-run-it checker.
Questions
- Can an RTX 5070 run a 70B model?
- Not at Q4_K_M in one card. Llama 3.3 70B needs about 45.8 GB at 8k context and this card can address 12 GB. Your options are a smaller model, a harsher quantisation with very little context, splitting across two cards, or renting a larger one for the hours you need it.
- What is the best model to run on an RTX 5070?
- Gemma 4 12B. At Q4_K_M and 8k context it needs about 8.2 GB of the 12 GB available and generates an estimated 68 tokens per second, which is faster than you can read. If quality matters more than size, Qwen3 8B fits at Q8 in about 10.7 GB.
- How many tokens per second does an RTX 5070 generate?
- It depends entirely on the size of the model, because generating a token means reading every active weight out of memory. At 672 GB/s and 70% efficiency, this card produces an estimated 101 tokens per second on an 8B model at Q4, and about 55 on the largest model it holds, Qwen3 14B. Reading speed is roughly 5 to 8 words per second, so anything above 25 feels immediate. Treat 101 as a ceiling on 672 GB/s rather than a measurement; the higher a figure, the further real runtimes fall below it.
- Is 12 GB enough for running LLMs locally?
- It runs 11 of the 34 models tracked here at Q4_K_M with 8k of context, up to 14.8B parameters. The honest test is not the model list but the context: Gemma 4 12B on this card holds about 131,072 tokens before memory runs out.
- RTX 5070 or RTX 5060 Ti for local models?
- The RTX 5060 Ti holds more: 16 GB against 12 GB, so it runs 12 of these models to this card's 11. On speed, this card is faster: 672 GB/s against 448 GB/s, and bandwidth is what sets chat speed.
- RTX 5070 or RTX 5060 Ti 16 GB for AI?
- The 5070 reads memory at 672 GB/s against 448 GB/s, so it generates half again as fast on any model both can hold. The 5060 Ti holds 16 GB against 12 GB, so it runs models the 5070 cannot fit. If your models fit in 12 GB, the 5070; if you want room for larger ones, the 5060 Ti 16 GB.
- Is 12 GB on the RTX 5070 enough for local AI?
- For small models, yes: it holds what any 12 GB card holds, and reads it quickly. It is the same capacity as a 3060 or a 4070, so it runs out where they do, and larger models need a 16 GB or 24 GB card.
- Memory: weights (parameters × bytes per parameter) + KV cache (2 × layers × KV heads × head dim × context × bytes for a standard transformer; sliding-window, hybrid and latent-attention models counted as they actually cache) + 0.5 GB runtime + 4% of weights. Full derivation in the VRAM calculator.
- Speed: 70% × bandwidth ÷ active weight bytes, single stream, no batching, prompt processing excluded. A ceiling, not a measurement: the 3 of 11 figures marked ≤ are where real runtimes fall furthest below the number, because so few weights are read per token. See the tokens-per-second estimator.
- Card specification (manufacturer figures): 12 GB, 672 GB/s, 250 W; see NVIDIA's own page for this card. The bandwidth figure: NVIDIA's GeForce comparison page.
- Model architecture values come from each model's published config; the self-hosted LLM guide explains what each one changes.