RTX 4060 — what LLMs can it run?
44 of 110 indexed models fit comfortably in 8GB at 4K context, each at the highest-quality quant that still leaves headroom.
The short answer for RTX 4060
8 GB, 272 GB/s GDDR6. 44 of 110 models in this index fit comfortably at 4K context; 51 load at all.
- Biggest that fits
- Stable LM 2 12B Chat — 7.0 GB at AWQ INT4 — 1.0 GB spare at 4K, so longer context comes out of a thin margin.
- Room to grow
- Qwen3 4B Instruct — 4.0 GB at Q6_K — under 60% of the card, which leaves 4.0 GB for a long context window or a second process.
- Speed ceiling
- Llama 3.1 8B Instruct at Q4_K_M reads 4.6 GB of weights per token, so 272 GB/s puts a hard ceiling near 59 tok/s. That is arithmetic on two published numbers, not a benchmark — real throughput lands below it. No run on this card has been measured here, so there is nothing to compare the ceiling against.
- The wall you will hit
- Bandwidth, not capacity. The models on this page fit, but at 272 GB/s this card reads the whole weight set once per generated token — a 608 GB/s RTX 3070 Ti holds exactly the same 8 GB and moves those bytes 2.2× faster. Expect the same quant to generate proportionally slower here.
14B · 5
| Model | Quant | Est. VRAM | Headroom | tok/s |
|---|---|---|---|---|
| Stable LM 2 12B Chat12B | AWQ INT4 | 7.01 GB | +1 GB | ≤47Ceiling |
| Solar 10.7B Instruct11B | AWQ INT4 | 6.42 GB | +1.6 GB | ≤54Ceiling |
| Falcon 3 10B Instruct10B | GPTQ INT4 | 6.12 GB | +1.9 GB | ≤55Ceiling |
| Gemma 2 9B Instruct9B | AWQ INT4 | 6.27 GB | +1.7 GB | ≤62Ceiling |
| Qwen3.5 9B9B | Q4_K_M | 6.58 GB | +1.4 GB | ≤46Ceiling |
7B · 25
| Model | Quant | Est. VRAM | Headroom | tok/s |
|---|---|---|---|---|
| GLM-4-9B-Chat9B | Q4_K_M | 5.87 GB | +2.1 GB | ≤53Ceiling |
| Qwen3-VL 8B Instruct8B | Q4_K_M | 6.19 GB | +1.8 GB | ≤54Ceiling |
| Granite 4.2 8B8B | Q4_K_M | 6.28 GB | +1.7 GB | ≤54Ceiling |
| Ministral 3 8B Instruct8B | Q5_K_M | 6.92 GB | +1.1 GB | ≤47Ceiling |
| Qwen2-VL 7B Instruct7B | Q4_K_M | 5.49 GB | +2.5 GB | ≤57Ceiling |
| Granite 3.1 8B Instruct8B | Q4_K_M | 5.88 GB | +2.1 GB | ≤58Ceiling |
| Qwen3 8B Instruct8B | EXL2 4.65bpw | 5.59 GB | +2.4 GB | ≤60Ceiling |
| Llama 3.1 8B Instruct8B | EXL2 4.65bpw | 5.43 GB | +2.6 GB | ≤61Ceiling |
| Nous Hermes 3 Llama 3.1 8B8B | EXL2 4.65bpw | 5.43 GB | +2.6 GB | ≤61Ceiling |
| Aya 23 8B8B | Q4_K_M | 5.64 GB | +2.4 GB | ≤59Ceiling |
| OpenChat 3.6 8B8B | EXL2 4.65bpw | 5.43 GB | +2.6 GB | ≤61Ceiling |
| DeepSeek-R1-Distill-Llama-8B8B | Q5_K_M | 6.51 GB | +1.5 GB | ≤50Ceiling |
| InternLM2 7B Chat7B | Q4_K_M | 5.45 GB | +2.6 GB | ≤61Ceiling |
| Qwen2.5 7B Instruct7B | Q6_K | 6.77 GB | +1.2 GB | ≤46Ceiling |
| Qwen2.5-Coder 7B Instruct7B | EXL2 4.65bpw | 4.87 GB | +3.1 GB | ≤65Ceiling |
| DeepSeek-R1-Distill-Qwen-7B7B | EXL2 4.65bpw | 4.87 GB | +3.1 GB | ≤65Ceiling |
| Gemma 4 E4B ITE4B | Q4_K_M | 4.88 GB | +3.1 GB | ≤63Ceiling |
| OLMo 2 7B Instruct7B | Q4_K_M | 6.82 GB | +1.2 GB | ≤65Ceiling |
| OLMo 3 7B Instruct7B | Q4_K_M | 6.82 GB | +1.2 GB | ≤65Ceiling |
| WizardLM-2 7B7B | Q4_K_M | 5.14 GB | +2.9 GB | ≤65Ceiling |
| Mistral 7B Instruct v0.37B | Q6_K | 6.75 GB | +1.3 GB | ≤48Ceiling |
| Zephyr 7B Beta7B | Q6_K | 6.75 GB | +1.3 GB | ≤48Ceiling |
| Gemma 4 E2B ITE2B | Q8_0 | 5.2 GB | +2.8 GB | ≤58Ceiling |
| Qwen3.5 4B4B | Q8_0 | 4.97 GB | +3 GB | ≤62Ceiling |
| Gemma 3 4B IT4B | Q8_0 | 5.05 GB | +3 GB | ≤63Ceiling |
≤3B · 14
| Model | Quant | Est. VRAM | Headroom | tok/s |
|---|---|---|---|---|
| Qwen3 4B Instruct4B | Q6_K | 4.05 GB | +4 GB | ≤87Ceiling |
| Phi-4 Mini Instruct3.8B | Q8_0 | 4.81 GB | +3.2 GB | ≤70Ceiling |
| Phi-3.5 Mini Instruct3.8B | Q8_0 | 5.89 GB | +2.1 GB | ≤70Ceiling |
| Granite 4.2 3B3B | Q8_0 | 4.41 GB | +3.6 GB | ≤74Ceiling |
| Llama 3.2 3B Instruct3B | Q8_0 | 4.05 GB | +4 GB | ≤84Ceiling |
| Qwen2.5 3B Instruct3B | Q8_0 | 3.59 GB | +4.4 GB | ≤87Ceiling |
| SmolLM3 3B3B | Q8_0 | 3.73 GB | +4.3 GB | ≤87Ceiling |
| Gemma 2 2B Instruct2B | Q8_0 | 3.34 GB | +4.7 GB | ≤103Ceiling |
| Qwen3.5 2B2B | Q8_0 | 2.23 GB | +5.8 GB | ≤137Ceiling |
| Qwen3 1.7B Instruct1.7B | Q8_0 | 2.39 GB | +5.6 GB | ≤156Ceiling |
| Qwen2.5 1.5B Instruct1.5B | Q8_0 | 1.83 GB | +6.2 GB | ≤175Ceiling |
| Llama 3.2 1B Instruct1B | Q8_0 | 1.51 GB | +6.5 GB | ≤218Ceiling |
| Qwen3.5 0.8B0.8B | Q8_0 | 0.94 GB | +7.1 GB | ≤336Ceiling |
| Qwen2.5 0.5B Instruct0.5B | Q8_0 | 0.6 GB | +7.4 GB | ≤555Ceiling |
How this list is built
Each row is the lowest-perplexity-loss quant of that model whose estimated total — weights plus KV cache at 4K context plus activation buffer — uses at most 88% of the card. That is the calculator's "green" threshold, so every row here has real headroom rather than only just fitting. Raise the context length and the list shortens; the calculator lets you check any combination directly. The speed column is this card: a measured run, a range scaled from this card's own Llama 2 7B benchmark, or a bandwidth ceiling. A figure that only exists for the RTX 4090 is labelled as that card.
A further 7 models load but with no headroom to spare (up to 105% of VRAM) — the Quant Hub’s GPU chips count those too, which is why its number is higher.
Measured on this card
No benchmark runs in this index were recorded on a RTX 4060. Every figure on this page is calculated from the model architecture and the quant level — treat them as estimates, not measurements.
Cards with the same budget
What fits is decided by memory, so every 8GB card of this type returns the same list. These pages are not different answers — they differ in throughput, which this index does not measure per card.
Setup guides for this card
Step-by-step guides written for the RTX 4060's runtime or memory size, with the commands and the checks that prove the model is running on the card.
Stepping up
A RTX 3080 10G (10GB) fits 12 more of the indexed models than this card. RTX 3080 10G →
Common questions
What is the best local LLM for an RTX 4060?
For everyday use, Qwen3 4B Instruct at Q6_K — about 4.0 GB of the card's 8 GB at 4K context, leaving 4.0 GB for a longer window. If you want the largest thing that will load, that is Stable LM 2 12B Chat at AWQ INT4 (7.0 GB, 1.0 GB spare). "Best" here means best fit for the memory budget — this index does not run task benchmarks, so it cannot tell you which model is smarter.
Can an RTX 4060 run Mellum2.1 12B-A2.5B Thinking?
No. Its smallest build here, Q4_K_M, needs about 8.6 GB and this card has 8.0 GB usable — short by 0.6 GB before any context beyond 4K. The largest model this card does clear is Stable LM 2 12B Chat.
How many tokens per second does an RTX 4060 do on an 8B model at Q4?
This index has no measured run on this card, so it does not publish a figure. What can be stated from specifications: generating a token requires reading every weight once, Llama 3.1 8B Instruct at Q4_K_M is 4.6 GB of weights, and this card moves 272 GB/s — a ceiling near 59 tok/s. Batch size, context length, the runtime and how much of the model sits in cache all take you below it.
RTX 4060 or RTX 3070 Ti for local LLMs?
They hold the same models: 8 GB against 8 GB fits 44 and 44 of 110 respectively at 4K. The difference is throughput — 272 GB/s against 608 GB/s, a 2.2× gap in how fast the weights can be read, which is what token generation is bound by. The RTX 3070 Ti generates faster on any model both can hold.
44 of 110 indexed models fit comfortably in 8GB at 4K context, each at the highest-quality quant that still leaves headroom.
This page's figures change when the model or the runtime does.Last updated 2026-10-18 RSS → /feed.xml