NVIDIA GeForce RTX 5090
32 GB · 1,792 GB/s · 575 W
In short
The NVIDIA GeForce RTX 5090 has 32 GB of GDDR7 memory with 1,792 GB/s of bandwidth, which sets how fast it generates text. At 32k context it runs 71 of the 128 open models we track fully in memory; the most capable is Qwen 3.8 27B at Q8_0, at about 35 tok/s (estimated). The best new price for the NVIDIA GeForce RTX 5090 in the US is $4,921 at eBay; used, it starts at $6,450 (asking price).
Updated . Speeds are estimates; how they are worked out.
Prices
Every retailer we track in the US, cheapest first. Buy links may earn us a commission.
| Retailer | Condition | Price | Stock | Checked | Buy |
|---|---|---|---|---|---|
| eBayGIGABYTE AORUS GeForce RTX 5090 32GB GDDR7 PCI Express 5.0 ATX Graphics Card GV- | New | $4,921 | In stock | 1 h ago | Buy |
| eBayGIGABYTE AORUS GeForce RTX 5090 32GB GDDR7 PCI Express 5.0 ATX Graphics Card GV- | New | $5,654 | In stock | 4 h ago | Buy |
| eBayASUS TUF Gaming GeForce RTX 5090 32GB GDDR7 Graphics Card | used, asking price | $6,450 | In stock | 4 h ago | Buy |
| eBayPNY NVIDIA GeForce RTX 5090 OC 32GB GDDR7 PCIe 5.0 x16 Graphics Card excellent | used, asking price | $6,399+ $59 delivery | In stock | 4 h ago | Buy |
| eBayMSI GeForce RTX 5090 32GB VENTUS 3X OC GDDR7 Graphics Card | used, asking price | $6,749 | In stock | 4 h ago | Buy |
| eBayASUS TUF GeForce RTX 5090 Graphics Card 32GB VRAM | used, asking price | $6,700+ $50 delivery | In stock | 1 h ago | Buy |
| eBayPalit GeForce RTX 5090 GameRock 32GB GDDR7 PCI-Express Graphics Card- | used, asking price | $6,775 | In stock | 4 h ago | Buy |
| eBayGigabyte AORUS GeForce RTX 5090 MASTER ICE | used, asking price | $6,900 | In stock | 4 h ago | Buy |
| eBayGigabyte AORUS GeForce RTX 5090 Master 32GB GDDR7 GPU Video Card | used, asking price | $6,949 | In stock | 4 h ago | Buy |
| eBayMSI GeForce RTX 5090 SUPRIM LIQUID SOC 32GB GDDR7 PCI-E 5.0 Graphics Card GPU | used, asking price | $6,949 | In stock | 4 h ago | Buy |
| eBayMSI GeForce RTX 5090 32GB GDDR7 VENTUS 3X OC Graphics Card | used, asking price | $7,000+ $70 delivery | In stock | 4 h ago | Buy |
| eBayASUS TUF GeForce RTX 5090 32GB GDDR7 Gaming OC Edition PCIe 5.0 Graphics Card | New | $7,199 | In stock | 1 h ago | Buy |
| eBayNEW ASUS ProArt GeForce RTX 5090 32GB GDDR7 OC Edition Graphics Card | used, asking price | $7,199 | In stock | 4 h ago | Buy |
| eBayASUS ROG Astral RTX 5090 OC BTF Edition 32GB GDDR7 | New | $7,399 | In stock | 1 h ago | Buy |
| eBayGIGABYTE NVIDIA GeForce RTX 5090 GAMING OC 32GB GDDR7 | New | $7,400 | In stock | 4 h ago | Buy |
| eBayASUS TUF Gaming GeForce RTX 5090 32GB Gaming Graphics Card | New | $7,400+ $30 delivery | In stock | 1 h ago | Buy |
| eBayASUS TUF GeForce RTX 5090 32GB GDDR7 Gaming OC Edition PCIe 5.0 Graphics Card | New | $7,450+ $30 delivery | In stock | 1 h ago | Buy |
| eBayGIGABYTE AORUS RTX 5090 MASTER ICE OC 32GB White GV-N5090AORUSM ICE-32GD | used, asking price | $7,499 | In stock | 4 h ago | Buy |
| eBayGigabyte AORUS RTX 5090 Infinity OC 32GB GDDR7 PCIe 5.0 Dual-Fan GPU NEW TESTED | New | $7,499 | In stock | 4 h ago | Buy |
| eBayASUS TUF Gaming GeForce RTX 5090 32GB GDDR7 PCIe 5.0 HDMI/DP | New | $7,500 | In stock | 4 h ago | Buy |
| eBay🔥ASUS RTX 5090 TUF 32GB NVIDIA GeForce Graphics Card | New | $7,500 | In stock | 4 h ago | Buy |
| eBayASUS ROG Astral OC NVIDIA GeForce RTX 5090 32GB GDDR7 | New | $7,500+ $40 delivery | In stock | 4 h ago | Buy |
| eBayGigabyte AORUS RTX 5090 Master ICE 32GB White GV-N5090AORUSM New | New | $7,600 | In stock | 4 h ago | Buy |
| eBay🔥ASUS RTX 5090 TUF OC 32GB NVIDIA GeForce Graphics Card | New | $7,640 | In stock | 4 h ago | Buy |
| eBayUsed GeForce RTX 5090 on eBay | used, asking price | $7,688 | In stock | 6 h ago | Buy |
| eBay🔥ASUS RTX 5090 ROG Astral OC 32GB NVIDIA GeForce | New | $7,700 | In stock | 4 h ago | Buy |
| AmazonMSI GeForce RTX 5090 32G SUPRIM SOC | New | $8,000 | In stock | 3 h ago | Buy |
| eBayNVIDIA RTX 5090 32GB 575W pulled from NEW DELL Alienware Area-51 0 Hours | New | $8,395 | In stock | 4 h ago | Buy |
| AmazonZOTAC Gaming Geforce RTX 5090 Amp Extreme Infinity Nvidia 32 Gb | New | $8,479 | In stock | 19 min ago | Buy |
| AmazonASUS GeForce RTX 5090 Astral LC Overclocked Liquid Cooled 32GB GDDR7 PCIe 5.0… | New | $8,800+ $28 delivery | In stock | 3 h ago | Buy |
| AmazonGigabyte AORUS GeForce RTX 5090 Infinity 32G Graphics Card | New | $9,000 | In stock | 19 min ago | Buy |
| AmazonASUS TUF Gaming GeForce RTX 5090 Triple Fan GPU | New | $9,300 | In stock | 19 min ago | Buy |
| AmazonASUS ROG Astral GeForce RTX 5090 32GB GDDR7 White OC Edition Gaming Quad Fan… | New | $9,500 | In stock | 3 h ago | Buy |
| AmazonMSI GeForce RTX 5090 32G Vanguard SOC Launch Edition Graphics Card VD8998 | New | $9,979 | In stock | 3 h ago | Buy |
| AmazonASUS ROG Astral GeForce RTX 5090 Edition 20 OC Quad-Fan Gaming Graphics Card | New | $11,500 | In stock | 3 h ago | Buy |
| eBayASUS ROG Astral GeForce RTX 5090 32GB GDDR7 NEW UNUSED Graphics Card | used, asking price | $6,999 | Out of stock | 4 h ago | Buy |
| eBayAORUS Master Ice NVIDIA GeForce RTX 5090 Graphics Card | used, asking price | $7,000 | Out of stock | 4 h ago | Buy |
| eBayNVIDIA GeForce RTX 5090 Founders Edition Graphics Card | used, asking price | $7,000 | Out of stock | 4 h ago | Buy |
| eBayNVIDIA GeForce RTX 5090 FE Founders Edition 32GB Graphics Card | used, asking price | $7,250 | Out of stock | 4 h ago | Buy |
| eBayASUS GeForce RTX 5090 ROG Astral OC Edition 32GB GDDR7- NEW/SEALED | New | $7,500 | Out of stock | 4 h ago | Buy |
| eBayASUS TUF GeForce RTX 5090 32GB GDDR7 Gaming OC Edition PCIe 5.0 Graphics Card | New | $7,599 | Out of stock | 4 h ago | Buy |
Used on eBay: median asking price $7,688, most between $7,000 and $9,000 (92 listings, 2026-10-30). Used price history
Price history
Lowest price shown each day over the last 90 days.
What two of them run
Every model in the catalogue, with the best quant for this hardware. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @32k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| Qwen 3.8 27B Alibaba Qwen | Q8_0 | fits | 30.5 GB | 256k | 35tok/sest. | 1517 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 122B-A10B Alibaba Qwen | Q3_K_M | fits | 54.7 GB | 256k | 122tok/sest. | 3557 tok/s | chat, coding, agents, long documents |
| Qwen 3.6 27B Alibaba Qwen | Q8_0 | fits | 30.1 GB | 256k | 36tok/sest. | 1517 tok/s | chat, coding, agents, long documents |
| Qwen3 Coder Next Alibaba Qwen | Q5_K_M | fits | 55.1 GB | 256k | 183tok/sest. | 8573 tok/s | chat, coding, agents, long documents |
| gpt-oss 120B OpenAI | MXFP4 | fits | 61.5 GB | 64k108k with q8 cache | 154tok/sest. | 5745 tok/s | chat, coding, agents, long documents |
| Nemotron 3 Super 120B-A12B NVIDIA | UD-Q3_K_M | fits | 59.2 GB | 256k | 104tok/sest. | 3597 tok/s | chat, coding, agents, long documents |
| Gemma 4 31B | Q8_0 | fits | 35.0 GB | 256k | 31tok/sest. | 1351 tok/s | chat, coding, agents, long documents |
| Mistral Small 4 119B Mistral AI | UD-Q3_K_M | fits | 52.6 GB | 397k630k with q8 cache | 168tok/sest. | 3954 tok/s | chat, coding, agents, long documents |
| Qwen 3.6 35B-A3B Alibaba Qwen | Q8_0 | fits | 36.3 GB | 256k | 175tok/sest. | 10830 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 27B Alibaba Qwen | Q8_0 | fits | 30.1 GB | 256k | 36tok/sest. | 1517 tok/s | chat, coding, agents, long documents |
| Gemma 4 26B-A4B | Q8_0 | fits | 27.1 GB | 256k | 132tok/sest. | 9635 tok/s | chat, coding, agents, long documents |
| GLM 4.5 Air Z.ai | Q3_K_M | fits | 60.3 GB | 45k83k with q8 cache | 50tok/sest. | 1469 tok/s | chat, coding, agents |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q5_K_M | fits | 55.0 GB | 256k | 224tok/sest. | 10199 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q8_0 | fits | 36.3 GB | 256k | 175tok/sest. | 10830 tok/s | chat, coding, agents, long documents |
| GLM 4.7 Flash Z.ai | Q8_0 | fits | 32.6 GB | 198k | 95tok/sest. | 5654 tok/s | chat, coding, agents, long documents |
| Devstral Small 2 24B Mistral AI | Q8_0 | fits | 29.6 GB | 236k384k with q8 cache | 37tok/sest. | 1601 tok/s | chat, coding, agents, long documents |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | UD-Q6_K_XL | fits | 34.1 GB | 256k | 187tok/sest. | 12358 tok/s | chat, coding, agents, long documents |
| gpt-oss 20B OpenAI | MXFP4 | fits | 13.3 GB | 128k | 206tok/sest. | 8085 tok/s | chat, coding, agents, long documents |
| Kimi Linear 48B-A3B Moonshot AI | Q8_0 | fits | 50.2 GB | 959k1024k with q8 cache | 191tok/sest. | 9333 tok/s | chat, coding, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q8_0 | fits | 34.5 GB | 256k | 104tok/sest. | 4980 tok/s | chat, coding, agents, long documents |
| Qwen3 32B Alibaba Qwen | Q8_0 | fits | 41.7 GB | 40k | 26tok/sest. | 942 tok/s | chat, coding |
| Gemma 4 12B | Q8_0 | fits | 13.9 GB | 256k | 82tok/sest. | 3518 tok/s | chat, coding, long documents |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 11.2 GB | 256k | 104tok/sest. | 4384 tok/s | chat, coding, long documents |
| Mistral Small 3.2 24B Mistral AI | Q8_0 | fits | 29.6 GB | 128k | 37tok/sest. | 1601 tok/s | chat, coding, long documents |
| Gemma 3 27B | Q8_0 | fits | 30.9 GB | 128k | 35tok/sest. | 1610 tok/s | chat, coding, long documents |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q8_0 | fits | 41.7 GB | 114k128k with q8 cache | 26tok/sest. | 1081 tok/s | chat, coding |
| Llama 3.3 70B Meta | Q5_K_M | fits | 57.8 GB | 47k88k with q8 cache | 18tok/sest. | 511 tok/s | chat |
| Ministral 3 14B Mistral AI | Q8_0 | fits | 19.7 GB | 256k | 56tok/sest. | 2435 tok/s | chat, coding, long documents |
| Qwen3 14B Alibaba Qwen | Q8_0 | fits | 20.9 GB | 40k | 53tok/sest. | 2189 tok/s | chat, coding, long documents |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q8_0 | fits | 13.9 GB | 128k | 82tok/sest. | 3611 tok/s | chat, coding, long documents |
| Qwen3 8B Alibaba Qwen | Q8_0 | fits | 13.9 GB | 40k | 82tok/sest. | 3611 tok/s | chat, coding, long documents |
| Phi-4 14B Microsoft | Q8_0 | fits | 22.0 GB | 16k | 50tok/sest. | 2201 tok/s | chat, coding, long documents |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 6.5 GB | 256k | 198tok/sest. | 8202 tok/s | chat, coding, long documents |
| Ministral 3 8B Mistral AI | Q8_0 | fits | 13.9 GB | 256k | 82tok/sest. | 3488 tok/s | chat, coding, long documents |
| Granite 4.2 8B IBM | Q8_0 | fits | 15.0 GB | 128k | 76tok/sest. | 3321 tok/s | chat, coding, agents, long documents |
| Gemma 4 E4B | Q8_0 | fits | 9.8 GB | 128k | 121tok/sest. | 5507 tok/s | chat, coding, long documents |
| LFM2.5 8B-A1B Liquid AI | Q8_0 | fits | 11.2 GB | 125k | 249tok/sest. | 18013 tok/s | chat, coding, long documents |
| Llama 3.1 8B Meta | Q6_K | fits | 11.4 GB | 128k | 102tok/sest. | 3815 tok/s | chat, coding, long documents |
| Qwen2.5 0.5B Alibaba Qwen | Q8_0 | fits | 2.2 GB | 32k | 1190tok/sest. | 39231 tok/s | chat, coding, long documents |
| Qwen3.5 0.8B Alibaba Qwen | Q8_0 | fits | 2.4 GB | 256k | 901tok/sest. | 36851 tok/s | chat, coding, agents, long documents |
| Qwen3.5 2B Alibaba Qwen | Q5_K_M | fits | 3.0 GB | 256k | 598tok/sest. | 17568 tok/s | chat, coding, agents, long documents |
| LFM3 1.7B Liquid AI | Q8_0 | fits | 3.4 GB | 256k | 492tok/sest. | 21869 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 1.5B DeepSeek | Q8_0 | fits | 3.9 GB | 128k | 392tok/sest. | 14758 tok/s | chat, coding, long documents |
| MiniCPM5 2B openbmb | Q8_0 | fits | 5.1 GB | 128k | 272tok/sest. | 8814 tok/s | chat, coding, agents, long documents |
| Spark X2.5 4B XHToken | mlx-4bit | fits | 5.3 GB | 1024k | 258tok/sest. | 8655 tok/s | chat, coding, agents, long documents |
| LFM2.5 2.6B Liquid AI | Q8_0 | fits | 5.8 GB | 128k | 227tok/sest. | 9983 tok/s | chat, coding, agents, long documents |
| LFM2.5 VL 3B Liquid AI | Q8_0 | fits | 5.8 GB | 32k | 227tok/sest. | 9155 tok/s | chat, coding, agents, long documents |
| Mellum2.1 12B A2.5B Thinking JetBrains | Q8_0 | fits | 13.8 GB | 128k | 210tok/sest. | 13006 tok/s | chat, coding, agents, long documents |
| FrogNano 4B 2609 Microsoft | Q8_0 | fits | 6.6 GB | 256k | 193tok/sest. | 8202 tok/s | chat, coding, agents, long documents |
| granite vision 4.1 4b IBM | Q8_0 | fits | 7.2 GB | 128k | 176tok/sest. | 6396 tok/s | chat, coding, long documents |
| granite 4.2 3b IBM | Q8_0 | fits | 7.4 GB | 128k | 169tok/sest. | 6703 tok/s | chat, coding, agents, long documents |
| Kimi VL A3B Thinking 2506 Moonshot AI | Q8_0 | fits | 18.0 GB | 128k | 162tok/sest. | 9773 tok/s | chat, coding, long documents |
| Kimi VL A3B Moonshot AI | mlx-8bit | fits | 18.8 GB | 128k | 156tok/sest. | 8224 tok/s | chat, coding, agents, long documents |
| Ministral 3 3B Instruct 2512 Mistral AI | Q8_0 | fits | 7.9 GB | 256k | 156tok/sest. | 6408 tok/s | chat, coding, agents, long documents |
| LFM2.5 24B A2B Liquid AI | Q8_0 | fits | 27.5 GB | 128k | 139tok/sest. | 10039 tok/s | chat, coding, agents, long documents |
| GLM 4 9B 0414 Z.ai | mlx-5bit | fits | 10.2 GB | 32k | 116tok/sest. | 3184 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 7B DeepSeek | Q8_0 | fits | 10.6 GB | 128k | 111tok/sest. | 4313 tok/s | chat, coding, long documents |
| GLM 4.6V Flash Z.ai | Q8_0 | fits | 11.9 GB | 128k | 98tok/sest. | 3003 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Llama 8B DeepSeek | Q8_0 | fits | 13.2 GB | 128k | 87tok/sest. | 3815 tok/s | chat, coding, long documents |
| GLM 4.1V 9B Thinking Z.ai | mlx-8bit | fits | 13.5 GB | 64k | 85tok/sest. | 3003 tok/s | chat, coding, long documents |
| Ministral 3 8B Reasoning 2512 Mistral AI | Q8_0 | fits | 13.9 GB | 256k | 82tok/sest. | 3488 tok/s | chat, coding, agents, long documents |
| Ministral 3 14B Reasoning 2512 Mistral AI | Q8_0 | fits | 19.7 GB | 256k | 56tok/sest. | 2435 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 14B DeepSeek | Q8_0 | fits | 21.9 GB | 128k | 50tok/sest. | 2060 tok/s | chat, coding, long documents |
| Qwen3.8 27B OBLITERATED OBLITERATUS | mlx-4bit | fits | 22.7 GB | 256k | 48tok/sest. | 1517 tok/s | chat, coding, agents, long documents |
| Devstral Small 2507 Mistral AI | Q8_0 | fits | 29.6 GB | 128k | 37tok/sest. | 1625 tok/s | chat, coding, agents, long documents |
| Magistral Small 2506 Mistral AI | Q8_0 | fits | 29.6 GB | 40k | 37tok/sest. | 1625 tok/s | chat, coding, long documents |
| Magistral Small 2507 Mistral AI | Q8_0 | fits | 29.6 GB | 40k | 37tok/sest. | 1625 tok/s | chat, coding, agents, long documents |
| Magistral Small 2509 Mistral AI | Q8_0 | fits | 29.6 GB | 128k | 37tok/sest. | 1601 tok/s | chat, coding, agents, long documents |
| Muse Glimmer 30B meta-models | Q8_0 | fits | 31.9 GB | 128k | 34tok/sest. | 1462 tok/s | chat, coding, agents, long documents |
| granite 4.2 30b IBM | Q8_0 | fits | 38.3 GB | 127k128k with q8 cache | 28tok/sest. | 1242 tok/s | chat, coding, long documents |
| DeepSeek R1 Distill Llama 70B DeepSeek | Q5_K_M | fits | 57.8 GB | 47k88k with q8 cache | 18tok/sest. | 511 tok/s | chat |
| Qwen 3.8 Flash Next Alibaba Qwen | UD-Q2_K_XL | offload | 75.6 GB | – | 5tok/sest. | 659 tok/s | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | offload | 79.2 GB | – | 11tok/sest. | 1748 tok/s | chat |
| GLM 4.5V Z.ai | Q4_K_M | offload | 66.3 GB | 13k24k with q8 cache | 22tok/sest. | 1387 tok/s | chat, coding |
| GLM 4.6V Z.ai | Q4_K_M | offload | 72.6 GB | – | 12tok/sest. | 1315 tok/s | chat |
| Kimi Dev 72B Moonshot AI | Q6_K | offload | 71.2 GB | 5k10k with q8 cache | 4tok/sest. | 463 tok/s | – |
| Devstral 2 123B Instruct 2512 Mistral AI | Q4_K_M | offload | 82.0 GB | – | 2tok/sest. | 251 tok/s | – |
| Mistral Medium 3.5 128B Mistral AI | Q4_K_M | offload | 82.0 GB | – | 2tok/sest. | 247 tok/s | – |
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | won't run | 94.2 GB | – | – | – | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | won't run | 103.0 GB | – | – | – | – |
| DeepSeek R1 DeepSeek | Q4_K_M | won't run | 380.1 GB | – | – | – | – |
| DeepSeek R1 0528 DeepSeek | Q4_K_M | won't run | 380.6 GB | – | – | – | – |
| DeepSeek R1 Zero DeepSeek | Q4_K_M | won't run | 380.1 GB | – | – | – | – |
| DeepSeek V3 DeepSeek | Q4_K_M | won't run | 380.1 GB | – | – | – | – |
| DeepSeek V3 0324 DeepSeek | Q4_K_M | won't run | 380.1 GB | – | – | – | – |
| DeepSeek V3.1 DeepSeek | Q4_K_M | won't run | 381.0 GB | – | – | – | – |
| DeepSeek V3.1 Terminus DeepSeek | Q4_K_M | won't run | 381.0 GB | – | – | – | – |
| DeepSeek V3.2 DeepSeek | Q4_K_M | won't run | 381.0 GB | – | – | – | – |
| DeepSeek V4 Flash 0731 DeepSeek | MXFP4 | won't run | 149.6 GB | – | – | – | – |
| DeepSeek V4 Flash Vision Exp DeepSeek | MXFP4 | won't run | 158.4 GB | – | – | – | – |
| GLM 4.5 Z.ai | Q4_K_M | won't run | 214.4 GB | – | – | – | – |
| GLM 4.6 Z.ai | Q4_K_M | won't run | 213.6 GB | – | – | – | – |
| GLM 4.7 Z.ai | Q4_K_M | won't run | 214.4 GB | – | – | – | – |
| GLM 5 Z.ai | MXFP4 | won't run | 386.4 GB | – | – | – | – |
| GLM 5.1 Z.ai | MXFP4 | won't run | 424.0 GB | – | – | – | – |
| GLM 5.2 Z.ai | UD-Q4_K_M | won't run | 437.9 GB | – | – | – | – |
| GLM 5.3 Z.ai | Q8_0 | won't run | 750.4 GB | – | – | – | – |
| Kimi K2 Moonshot AI | Q4_K_M | won't run | 581.6 GB | – | – | – | – |
| Kimi K2 Instruct 0905 Moonshot AI | Q4_K_M | won't run | 582.0 GB | – | – | – | – |
| Kimi K2 Thinking Moonshot AI | Q4_K_M | won't run | 582.0 GB | – | – | – | – |
| Kimi K2.5 Moonshot AI | Q4_K_M | won't run | 582.0 GB | – | – | – | – |
| MiniMax M2 MiniMax | Q4_K_M | won't run | 137.9 GB | – | – | – | – |
| MiniMax M2.1 MiniMax | Q4_K_M | won't run | 137.9 GB | – | – | – | – |
| MiniMax M2.5 MiniMax | MXFP4 | won't run | 124.3 GB | – | – | – | – |
| MiniMax M3 MiniMax | MXFP4 | won't run | 244.7 GB | – | – | – | – |
Specifications
From the manufacturer's published specifications; sources are linked where we have them.
- GPU
- GeForce RTX 5090
- Memory
- 32 GB GDDR7
- Memory bus
- 512-bit
- Memory bandwidth
- 1,792 GB/s
- FP16 compute
- 209 TFLOPS
- Software
- CUDA
- Board power
- 575 W
- PCIe generation
- Gen 5
- PCIe lanes
- x16
Compare with
Questions
- How much memory does the NVIDIA GeForce RTX 5090 have for AI models?
- The NVIDIA GeForce RTX 5090 has 32 GB of GDDR7 memory with 1,792 GB/s of bandwidth, which sets how fast it generates text.
- What is the best model the NVIDIA GeForce RTX 5090 can run locally?
- Qwen 3.8 27B (Q8_0) fits fully in its memory at 32k context and generates at about 35 tok/s, an estimate from its memory bandwidth. 71 of the 128 models we track fit at that context.
- Can the NVIDIA GeForce RTX 5090 run DeepSeek V4 Flash?
- No: even at UD-Q2_K_XL, DeepSeek V4 Flash needs about 94 GB at 32k context, more than the NVIDIA GeForce RTX 5090 and system RAM can hold.
- How much does the NVIDIA GeForce RTX 5090 cost in the US?
- The best new price for the NVIDIA GeForce RTX 5090 in the US is $4,921 at eBay; used, it starts at $6,450 (asking price).
Speeds on this page are for 32k context. See the best hardware for local AI at every budget.