AI models on a 16 GB graphics card
Open chat models judged for a PC card with 16 GB of video memory: which fit with every part on the card, the size to download, and what each one takes.
3341 of the 3940 chat models people download as GGUF fit a 16 GB card whole. The best-rated is Gemma 4 26B at UD-Q3_K_XL, which takes 14 GB. The largest rated model that fits is Qwen3.8 27B (27B).
Cards with 16 GB: GeForce RTX 5070 Ti, GeForce RTX 4080, GeForce RTX 4060 Ti 16GB, GeForce RTX 5080, GeForce RTX 5060 Ti 16GB, GeForce RTX 5080 Laptop, GeForce RTX 4080 Super, GeForce RTX 4070 Ti Super, GeForce RTX 4090 Laptop, RTX A4000, Radeon RX 9070 XT, Radeon RX 9070, Radeon RX 9060 XT 16GB, Radeon RX 7800 XT, Radeon RX 7600 XT, Radeon RX 6900 XT, Radeon RX 6800 XT, Radeon RX 6800, Arc A770 16GB. The same answer holds for any card with 16 GB of its own memory.
The best-rated models that fit
The 25 with the highest LM Arena rating (data, CC BY 4.0) that fit at 4 bits a weight or more, each at the most bits that fits. Below 4 bits a model's rating overstates it. Each model's page hands it to LM Studio, Jan, Ollama and llama.cpp.
| Model | Size | Fit | Takes | Longest chat | Arena | Arena coding |
|---|---|---|---|---|---|---|
| Gemma 4 26BGoogle · 25B, 3.8B active | UD-Q3_K_XL13 GB · unsloth | Fits with room | 14 GB | 32K | 1438 | 1482 |
| Qwen3.8 27BAlibaba Qwen · 27B | UD-Q4_K_S15 GB · unsloth | Fits | 16 GB | 8K | 1436 | 1498 |
| Qwen3.5 27BAlibaba Qwen · 27B | IQ4_XS15 GB · unsloth | Fits | 16 GB | 16K | 1409 | 1451 |
| GLM 4.7 FlashZ.ai · 30B, 3.9B active | Q3_K_M15 GB · unsloth | Fits | 16 GB | 32K | 1365 | 1423 |
| Gemma 3 27BGoogle · 27B | Q3_K_XL15 GB · bartowski | Fits | 16 GB | 8K | 1365 | 1357 |
| Gemma 3 12BGoogle · 12B | Q8_013 GB · lm-kit | Fits with room | 13 GB | 16K | 1342 | 1316 |
| Gemma 3n E4BGoogle · 6.9B | Q8_07.4 GB · unsloth | Fits with room | 7.7 GB | 32K | 1318 | 1307 |
| GPT-OSS 20BOpenAI · 21B, 4.2B active | UD-Q4_K_XL12 GB · unsloth | Fits with room | 14 GB | 128K | 1317 | 1368 |
| Granite 4.1 8BIBM · 8.8B | Q8_09.3 GB · ibm-granite | Fits with room | 13 GB | 32K | 1304 | 1353 |
| Gemma 3 4BGoogle · 3.9B | Q6_K_L3.4 GB · bartowski | Fits with room | 5.1 GB | 128K | 1303 | 1273 |
| Granite 4.2 8BIBM · 8.8B | Q8_09.3 GB · ibm-granite | Fits with room | 13 GB | 32K | 1290 | 1370 |
| Granite 4.2 3BIBM · 3.7B | Q8_03.9 GB · ibm-granite | Fits with room | 10 GB | 128K | 1290 | 1378 |
| Gemma 2 27BGoogle · 27B | Q3_K_M13 GB · bartowski | Fits | 15 GB | 8K | 1289 | 1304 |
| Gemma 2 9BGoogle · 9.2B | Q8_09.8 GB · bartowski | Fits with room | 12 GB | 8K | 1267 | 1270 |
| Phi 4Microsoft · 15B | Q6_K12 GB · microsoft | Fits with room | 13 GB | 8K | 1256 | 1306 |
| Aya Expanse 8BCohere · 8B | Q8_08.5 GB · bartowski | Fits with room | 9.3 GB | 8K | 1223 | 1255 |
| Llama 3.1 Tulu 3 8BAi2 · 8B | Q8_08.5 GB · lmstudio-community | Fits with room | 14 GB | 64K | 1221 | |
| Llama 3.1 8BMeta · 8B | UD-Q8_K_XL11 GB · unsloth | Fits with room | 13 GB | 32K | 1211 | 1259 |
| Granite 3.1 8BIBM · 8.2B | Q8_08.7 GB · bartowski | Fits with room | 12 GB | 32K | 1209 | |
| Gemma 2 2BGoogle · 2.6B | Q8_02.8 GB · bartowski | Fits with room | 3.3 GB | 8K | 1200 | 1191 |
| Phi 3 Medium 4KMicrosoft · 14B | Q6_K11 GB · bartowski | Fits with room | 12 GB | 4K | 1198 | 1231 |
| Internlm2 5 20BShanghai AI Lab · 20B | Q4_K_L12 GB · bartowski | Fits with room | 13 GB | 8K | 1191 | 1249 |
| Granite 3.1 2BIBM · 2.5B | Q8_02.7 GB · bartowski | Fits with room | 9 GB | 128K | 1179 | 1249 |
| Llama 3.2 3BMeta · 3.2B | Q8_03.4 GB · unsloth | Fits with room | 12 GB | 128K | 1167 | 1176 |
| SOLAR 10.7B V1.0Upstage · 11B | Q8_011 GB · TheBloke | Fits with room | 12 GB | 4K | 1152 |
The chat models Outmake picks
17 of these 28 fit. Their memory is what llama.cpp's own calculator reported for each size, not an estimate.
| Model | Size | Fit | Takes |
|---|---|---|---|
| Qwen3.8 27BAlibaba Qwen · 28B | UD-IQ4_XS15 GB | Fits | 15 GB |
| Gemma 4 26BGoogle · 26B, 3.8B active | UD-Q3_K_M14 GB | Fits | 15 GB |
| K2-Horizon 7BInstitute of Foundation Models · 9B | Q6_K7.8 GB | Fits with room | 10 GB |
| K2-Horizon 3.7BInstitute of Foundation Models · 5.1B | Q6_K4.4 GB | Fits with room | 7.0 GB |
| Gemma 4 12BGoogle · 12B | Q6_K10.0 GB | Fits with room | 13 GB |
| GLM 4.7 FlashZ.ai · 31B, 3B active | Q3_K_M15 GB | Fits | 16 GB |
| Qwen3.5 9BAlibaba Qwen · 9B | Q6_K8.4 GB | Fits with room | 14 GB |
| Gemma 4 E4BGoogle · 8B | UD-Q5_K_XL7.6 GB | Fits with room | 7.6 GB |
| Devstral Small 2 24BMistral AI · 24B | Q3_K_M12 GB | Fits with room | 13 GB |
| Gemma 4 E2BGoogle · 5.1B | Q6_K5.5 GB | Fits with room | 5.5 GB |
| Qwen3.6 35BAlibaba Qwen · 36B, 3B active | UD-IQ3_S15 GB | Fits | 15 GB |
| Qwen3.5 4BAlibaba Qwen · 4B | Q6_K4.2 GB | Fits with room | 7.3 GB |
| GPT-OSS 20BOpenAI · 21B, 3.6B active | UD-Q8_K_XL13 GB | Fits with room | 13 GB |
| Qwen3.5 2BAlibaba Qwen · 2B | Q6_K2.2 GB | Fits with room | 3.6 GB |
| Qwen3.5 0.8BAlibaba Qwen · 0.8B | Q6_K0.8 GB | Fits with room | 2.4 GB |
| Gemma 4 31BGoogle · 31B | UD-IQ3_XXS13 GB | Fits | 16 GB |
| K2-Horizon 36BInstitute of Foundation Models · 37B, 4B active | Q2_K14 GB | Fits | 15 GB |
| Qwen3.5 122BAlibaba Qwen · 122B, 10B active | UD-Q3_K_XL58 GB | Bigger than the card | 62 GB |
| GPT-OSS 120BOpenAI · 117B, 5.1B active | UD-Q4_K_XL63 GB | Bigger than the card | 65 GB |
| Qwen3.8 Flash NextAlibaba Qwen · 177B, 6B active | UD-Q3_K_XL91 GB | Bigger than the card | 91 GB |
| MiniMax M2.7MiniMax · 229B, 10B active | UD-Q3_K_XL102 GB | Bigger than the card | 120 GB |
| Qwen3.5 397BAlibaba Qwen · 403B, 17B active | UD-IQ2_M124 GB | Bigger than the card | 129 GB |
| DeepSeek V4 FlashDeepSeek · 284B, 13B active | UD-Q3_K_XL128 GB | Bigger than the card | 129 GB |
| GLM 5.3Z.AI · 754B, 40B active | UD-Q2_K_XL254 GB | Bigger than the card | 265 GB |
| Kimi K2.6Moonshot AI · 1T, 32B active | UD-Q2_K_XL341 GB | Bigger than the card | 351 GB |
| Kimi K3Moonshot AI · 2.8T, 104B active | UD-IQ2_XXS712 GB | Bigger than the card | 716 GB |
| Qwen3.8 2.4TAlibaba Qwen · 2.4T, 95B active | UD-IQ2_XS731 GB | Bigger than the card | 736 GB |
| DeepSeek V4 ProDeepSeek · 1.6T, 49B active | UD-Q4_K_XL850 GB | Bigger than the card | 851 GB |
How this is worked out
A model fits when its weights, its memory of the chat so far and the engine's own share come to 15 GB or less: 92% of the card, with the rest left for the screen. "Fits with room" is under 80%. "Tight fit" is anything up to half a gigabyte short of the whole card: close other programs that draw with the card first.
"Bigger than the card" does not mean it won't run. llama.cpp and the apps built on it put what doesn't fit in ordinary memory and the model answers slower, often several times slower. So this page lists what fits whole.
There are no speeds here. Outmake's speed estimates are fitted to measurements on Apple chips, and no PC has been measured yet, so a number would be a guess. Memory was checked against the catalog of November 4, 2026; which cards have 16 GB against each maker's specification page on October 13, 2026.