Launch MSRPs, release cycles and deals windows as of November 9, 2026.
The NVIDIA RTX 5080 and NVIDIA RTX 5090 are both Enthusiast cards from NVIDIA's RTX 5000 generation, released within days of each other. It's a choice within one family rather than between ecosystems. Against the RTX 5080, the RTX 5090 costs $1,000 more at launch MSRP ($1,999 vs $999, 100% more), has 16 GB more VRAM (32 GB vs 16 GB) and draws 215W more at rated TDP (575W vs 360W).
Late in cycle — next generation likely within 12 months
Best for: 4K gamers and creators who want Blackwell performance without the 5090's price and power demands.
Full details →Mid-cycle — next generation may be on the horizon
Best for: Content creators, AI researchers, and enthusiast gamers who want the absolute fastest GPU regardless of price or power consumption.
Full details →| NVIDIA RTX 5080 | NVIDIA RTX 5090 | |
|---|---|---|
| Tier | Enthusiast | Enthusiast |
| Generation | RTX 5000 | RTX 5000 |
| VRAM | 16 GB | 32 GB |
| Memory Type | GDDR7 | GDDR7 |
| TDP | 360W | 575W |
| Upscaling | DLSS4 | DLSS4 |
| Ray Tracing | ✅ | ✅ |
| Launch MSRP | $999 | $1999 |
| Editor score | 8.6/10 | 9.4/10 |
| Released | Jan 30, 2025 | Jan 30, 2025 |
| Cycle length | ~850 days | ~850 days |
| Cycle advice | Wait | Caution |
| Deals advice | Buy | Caution |
| Successor | — | — |
Delivers excellent 4K frame rates at a lower TDP and price than the 5090 — the practical enthusiast choice.
Same DLSS 4 technology as the flagship, dramatically boosting frame rates in supported titles.
Runs on a 750W PSU comfortably, unlike the 5090's 1000W recommendation.
Double the VRAM of the RTX 5080 ensures headroom for 8K textures, AI model training, and multi-monitor setups.
Generates multiple frames per rendered frame, dramatically boosting perceived frame rates in supported games.
New shader cores, enhanced RT cores, and Tensor cores deliver the largest generational leap NVIDIA has shipped.
Released Jan 30, 2025, the RTX 5080 has been on the market for 21 months. The next model on our radar: the RTX 5080 Super, expected CES 2027, January 2027 (rumoured) — roughly 58 days out if that date holds. That puts the RTX 5080 about 92% of the way through its current cycle, which our model rates Wait.
Released Jan 30, 2025, the RTX 5090 has been on the market for 21 months. The next model on our radar: the RTX 6090, expected 2028 (rumoured) — roughly 418 days out if that date holds. That puts the RTX 5090 about 61% of the way through its current cycle, which our model rates Caution.
On timing alone, the RTX 5090 is the safer buy right now: it rates Caution, against Wait for the RTX 5080. If you lean towards the RTX 5080, it may be worth waiting to see what the RTX 5080 Super brings.
At launch MSRP the RTX 5080 is $1,000 cheaper than the RTX 5090 ($999 vs $1,999, 50% less). Per gigabyte of VRAM that's $62.4 for the RTX 5080 and $62.5 for the RTX 5090 — effectively level.
Both cards are in the Black Friday / Cyber Monday 2026 deals window, which runs to Dec 5, 2026 and is rated Caution for each. For the RTX 5080: usually the best annual GPU deals, but analysts do not expect memory supply to normalise before late 2027, so discounts are unlikely to reach the $999 MSRP. A rumoured RTX 5080 Super is reported for CES 2027. Only buy at or near MSRP. For the RTX 5090: usually the best annual GPU deals, but analysts do not expect memory supply to normalise before late 2027, so discounts are unlikely to reach the $1,999 MSRP. Only buy at or near MSRP.
Next on both deals calendars: New Year clearance 2027, from Jan 2, 2027 to Jan 20, 2027.
Our pick: NVIDIA RTX 5090
Is the RTX 5090 worth $1,000 more? That's 100% on top of the RTX 5080's launch MSRP; on the data we track it also has 16 GB more VRAM (32 GB vs 16 GB), carries a higher editor score (9.4 vs 8.6) and has the better buy/wait signal (Caution vs Wait). The cost side: 215W more rated TDP to cool and power.
Our editor scores favour the RTX 5090 (9.4 vs 8.6), so that's our pick if the budget stretches — otherwise the RTX 5080 is the sensible saving.
Buy/Wait and deals ratings come from release-cycle and pricing data, not benchmarks. How we rate →
Where they're level: memory type (GDDR7) and release timing (launched the same week).