NVIDIA GeForce RTX 2060
6 GB decides what fits. 336 GB/s decides how fast it runs once it does.
Open in the calculator →Best models for 6 GB, on every card that size →
Find one: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.
- Largest popular model that fitsllama-7b6.7B params · Q4_K_M · needs 5.96 GB~45 tok/s Faster than you readOpen in the calculator →
- Best fast pickgemma-4-E4B-it8.0B params · Q4_K_M · needs 5.86 GB~76 tok/s Faster than you readOpen in the calculator →
- Most downloaded that fitsQwen3.5-4B4.7B params · Q4_K_M · needs 3.98 GB~85 tok/s Faster than you readOpen in the calculator →
At 8,192 tokens of context with the whole model in device memory. Speeds are estimates from memory bandwidth, not benchmarks run on this card.
87 of 320 fit entirely
87 of the 320 models the engine can size fit entirely in device memory at Q4_K_M where it is published, otherwise the nearest published format, and 8,192 tokens. 130 run with some layers on system memory (32 GB assumed), and 103 do not run at all.
Featured models · Q4_K_M at 8,192 tokens, including offload
| Model | Needs | Verdict | Decode | Calculator |
|---|---|---|---|---|
| Qwen3.5-2B2.3B parameters | 2.32 GB | fits | ~185 tok/s | Open → |
| NVIDIA-Nemotron-3-Nano-4B-BF164.0B parameters | 3.46 GB | fits | ~86 tok/s | Open → |
| Qwen3-VL-8B-Instruct8.8B parameters | 7.04 GB | 8 of 36 layers on system RAM (1.12 GB) | ~26 tok/s with offload | Open → |
| Qwen3.5-9B9.7B parameters | 7.05 GB | 6 of 32 layers on system RAM (1.11 GB) | ~32 tok/s with offload | Open → |
| gemma-4-12B-it12B parameters | 9.54 GB | 22 of 48 layers on system RAM (3.61 GB) | ~13 tok/s with offload | Open → |
| gpt-oss-20b21B parameters | 13.8 GB | 15 of 24 layers on system RAM (7.97 GB) | ~35 tok/s with offload | Open → |
| Qwen3.8-27B28B parameters | 18.6 GB | 48 of 64 layers on system RAM (12.8 GB) | ~5.1 tok/s with offload | Open → |
| Qwen3.6-27B28B parameters | 18.6 GB | 48 of 64 layers on system RAM (12.8 GB) | ~5.1 tok/s with offload | Open → |
| Kimi-Linear-48B-A3B-Instruct49B parameters | 30.4 GB | 23 of 27 layers on system RAM (25.1 GB) | ~38 tok/s with offload | Open → |
| Qwen3.8-Flash-Next180B parameters | 111.6 GB | needs 105.9 GB of system RAM; 32 GB assumed | does not run | Open → |
| DeepSeek-V4-Flash-Vision-Exp305B parameters | 188.1 GB | needs 182.9 GB of system RAM; 32 GB assumed | does not run | Open → |
| GLM-5.3-Flash321B parameters | 198.3 GB | needs 192.9 GB of system RAM; 32 GB assumed | does not run | Open → |
These examples are selected from prominent labs using the catalogue’s latest Hugging Face 30-day downloads and repository-creation freshness signal, with newer releases guaranteed a place. Offloaded rows assume 32 GB of system RAM, and a speed is only shown for a row that runs. Fitting in memory is not the same as loading: whether the runtime and version you have supports each architecture and format on this machine has not been tested here. Each “Open” link carries the same model, format, context and RAM into the calculator.
Fits entirely in NVIDIA GeForce RTX 2060 memory — 87 of 320 sized models
The most demanding model that fits is llama-7b at Q4_K_M: 5.96 GB of the 6 GB, leaving 0.04 GB spare.
Fully resident at 8,192 tokens (or the model’s own maximum, where that is shorter), offload off, at 336 GB/s. Each row is a run of the engine for this configuration; the rows start with current, prominent releases and “fits” is memory, not a tested runtime. Older or less prominent models remain available through this search and “Show all”.
Showing 40 of 87 models that fit.
| Model | Format | Needs | Spare | Decode | Downloads / 30d | Calculator |
|---|---|---|---|---|---|---|
| Qwen3.5-4BQwen · 4.7B params | Q4_K_M | 3.98 GB | 2.02 GB | ~85 tok/s | 7.8M | Open → |
| Qwen3.5-2BQwen · 2.3B params | Q4_K_M | 2.32 GB | 3.68 GB | ~185 tok/s | 4.9M | Open → |
| gemma-4-E4B-itGoogle · 8.0B params | Q4_K_M | 5.86 GB | 0.14 GB | ~76 tok/s | 4.4M | Open → |
| NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA · 4.0B params | Q4_K_M | 3.46 GB | 2.54 GB | ~86 tok/s | 3.4M | Open → |
| gemma-4-E2B-itGoogle · 5.1B params | Q4_K_M | 4.02 GB | 1.98 GB | ~200 tok/s | 3M | Open → |
| Qwen3.5-0.8BQwen · 873M params | Q4_K_M | 1.46 GB | 4.54 GB | ~406 tok/s | 2.6M | Open → |
| North-Micro-Vision-InstructCohereLabs · 2.5B params | Q4_K_M | 2.99 GB | 3.01 GB | ~122 tok/s | 180.7K | Open → |
| granite-4.1-3bIBM · 3.4B params | Q4_K_M | 3.72 GB | 2.28 GB | ~82 tok/s | 521.8K | Open → |
| LFM2.5-2.6BLiquidAI · 2.7B params | Q4_K_M | 2.61 GB | 3.39 GB | ~132 tok/s | 108.1K | Open → |
| Qwen3-VL-4B-InstructQwen · 4.4B params | Q4_K_M | 4.72 GB | 1.28 GB | ~62 tok/s | 3.5M | Open → |
| Qwen3-VL-2B-InstructQwen · 2.1B params | Q4_K_M | 3.05 GB | 2.95 GB | ~112 tok/s | 2.8M | Open → |
| granite-4.2-3bIBM · 3.7B params | Q4_K_M | 3.72 GB | 2.28 GB | ~82 tok/s | 50.6K | Open → |
| LFM2.5-230MLiquidAI · 230M params | Q4_K_M | 1.05 GB | 4.95 GB | ~922 tok/s | 88.6K | Open → |
| Qwen3-0.6BQwen · 752M params | Q4_K_M | 2.22 GB | 3.78 GB | ~172 tok/s | 29.7M | Open → |
| LFM2.5-VL-3BLiquidAI · 3.1B params | Q4_K_M | 2.85 GB | 3.15 GB | ~132 tok/s | 26.7K | Open → |
| Qwen3-4B-Instruct-2507Qwen · 4.0B params | Q4_K_M | 4.72 GB | 1.28 GB | ~62 tok/s | 3.7M | Open → |
| Ling-3.0-tinyinclusionAI · 7.9B params | Q4_K_M | 5.85 GB | 0.15 GB | ~266 tok/s | 17.7K | Open → |
| Qwen3-4BQwen · 4.0B params | Q4_K_M | 4.51 GB | 1.49 GB | ~62 tok/s | 7.8M | Open → |
| LFM2.5-350MLiquidAI · 354M params | Q4_K_M | 1.13 GB | 4.87 GB | ~595 tok/s | 69.9K | Open → |
| Hy-MT2-1.8BTencent · 2.0B params | Q4_K_M | 2.47 GB | 3.53 GB | ~137 tok/s | 28.9K | Open → |
| LFM2.5-1.2B-InstructLiquidAI · 1.2B params | Q4_K_M | 1.63 GB | 4.37 GB | ~220 tok/s | 119.2K | Open → |
| Qwen3-1.7BQwen · 2.0B params | Q4_K_M | 3.02 GB | 2.98 GB | ~112 tok/s | 3.1M | Open → |
| Nemotron-3.5-Content-SafetyNVIDIA · 4.3B params | Q4_K_M | 3.73 GB | 2.27 GB | ~82 tok/s | 14.1K | Open → |
| SmolLM3-3BHuggingFaceTB · 3.1B params | Q4_K_M | 3.46 GB | 2.54 GB | ~91 tok/s | 617K | Open → |
| Ministral-3-3B-Instruct-2512Mistral AI · 3.8B params | Q4_K_M | 3.82 GB | 2.18 GB | ~76 tok/s | 114.1K | Open → |
| Qwen3-1.7B-BaseQwen · 1.7B params | Q4_K_M | 3.02 GB | 2.98 GB | ~112 tok/s | 696K | Open → |
| Qwen2.5-VL-3B-InstructQwen · 3.8B params | Q4_K_M | 3.41 GB | 2.59 GB | ~103 tok/s | 2.4M | Open → |
| Qwen3-4B-BaseQwen · 4.0B params | Q4_K_M | 4.72 GB | 1.28 GB | ~62 tok/s | 624.7K | Open → |
| Qwen2.5-0.5B-InstructQwen · 494M params | Q4_K_M | 1.31 GB | 4.69 GB | ~534 tok/s | 8.7M | Open → |
| Qwen2.5-7B-InstructQwen · 7.6B params | Q4_K_M | 5.95 GB | 0.05 GB | ~47 tok/s | 8.4M | Open → |
| Qwen2.5-1.5B-InstructQwen · 1.5B params | Q4_K_M | 2.15 GB | 3.85 GB | ~188 tok/s | 7.4M | Open → |
| OLMo-2-0425-1Ballenai · 1.5B params · sized at 4,096 tokens, its maximum | Q4_K_M | 2.27 GB | 3.73 GB | ~169 tok/s | 404.2K | Open → |
| SmolLM3-3B-BaseHuggingFaceTB · 3.1B params | Q4_K_M | 3.46 GB | 2.54 GB | ~91 tok/s | 133.5K | Open → |
| DeepSeek-R1-Distill-Qwen-1.5BDeepSeek · 1.8B params | Q4_K_M | 2.15 GB | 3.85 GB | ~188 tok/s | 961.3K | Open → |
| Qwen2.5-3B-InstructQwen · 3.1B params | Q4_K_M | 3.21 GB | 2.79 GB | ~103 tok/s | 4.2M | Open → |
| SmolLM2-135M-InstructHuggingFaceTB · 135M params | Q4_K_M | 1.09 GB | 4.91 GB | ~829 tok/s | 1.8M | Open → |
| Phi-tiny-MoE-instructMicrosoft · 3.8B params · sized at 4,096 tokens, its maximum | Q4_K_M | 3.22 GB | 2.78 GB | ~268 tok/s | 71.7K | Open → |
| SmolLM2-135MHuggingFaceTB · 135M params | Q4_K_M | 1.09 GB | 4.91 GB | ~829 tok/s | 1.8M | Open → |
| Phi-4-mini-instructMicrosoft · 3.8B params | Q4_K_M | 4.66 GB | 1.34 GB | ~65 tok/s | 370.4K | Open → |
| Qwen2.5-Coder-7B-InstructQwen · 7.6B params | Q4_K_M | 5.95 GB | 0.05 GB | ~47 tok/s | 2.3M | Open → |
Sized at the format most people actually download, not at FP16. The catalogue ordered by downloads →
Find the NVIDIA GeForce RTX 2060: Amazon ↗ · eBay (new and used) ↗Store links may pay us a commission. They never decide which card is suggested — the memory arithmetic does.
The specification behind every figure
What the manufacturer publishes for this device, and the pages it was read from.
Manufacturer specification
| Memory scope | dedicated |
|---|---|
| Capacity | 6 GB |
| Published options | 6 GB |
| Bandwidth | 336 GB/s |
| Memory type | GDDR6 |
| Bus width | 192-bit |
| FP32 peak | Not published |
| Dense matrix peak | Not published without sparsity |
| Graphics Card Power | 160 W |
Source ledger
Caveats
- NVIDIA does not publish this card's memory bandwidth any more, so the figure comes from the TechPowerUp GPU Database (https://www.techpowerup.com/gpu-specs/geforce-rtx-2060.c3310), a third-party source, and is labelled as an estimate rather than a published specification.
- No dense matrix throughput is published for this device in a form this catalogue accepts, so no compute roof is priced and time to first token is withheld.
Published capacity is a hardware ceiling, not guaranteed free runtime memory. The calculator shows the runtime reserve separately rather than folding it into a single number.
Run the diagnostic on the NVIDIA GeForce RTX 2060 →