NVIDIA is reportedly planning a second RTX 5050 variant with 9 GB of memory instead of the standard 8 GB, according to leaker MEGAsizeGPU who has a decent track record on NVIDIA leaks. Unlike the existing RTX 5050, which uses 8 GB of 20 Gbps GDDR6 memory on a 128-bit bus for 320 GB/s of bandwidth, the new model is said to adopt 3 GB GDDR7 modules. The updated configuration would total 9 GB across a 96-bit memory bus. While the narrower bus reduces interface width, the move to 28 Gbps GDDR7 would raise total memory bandwidth to 336 GB/s, a roughly 5% increase, alongside a 12.5% boost in VRAM capacity. Core specifications are expected to remain unchanged. The card would continue to use the GB207 GPU, based on the Blackwell architecture, with 2560 CUDA cores and a 130 W TDP. No changes to clock speeds or core counts have been mentioned.

The report also claims NVIDIA is preparing adjustments higher up the stack. A revised GeForce RTX 5060 may transition to the larger GB205 die, currently used in the RTX 5070, in a cut-down configuration. That would reportedly reduce the GPU from 6144 CUDA cores and a 192-bit bus to 3840 cores and a 128-bit interface. Such a move would require board partners to adopt new PCBs compatible with GB205 and standard 8-pin power connectors. According to the leak, these changes are tied to supply constraints affecting lower-tier RTX 50-series GPUs, including the RTX 5050 and RTX 5060. The GeForce RTX 5060 12 GB, which appeared in earlier roadmaps, is now reportedly canceled.

At Bilibili World 2025, Colorful quietly introduced a unique all-white version of its iGame Ultra Duo SSD graphics card that combines high-end GPU performance with built-in storage. The card features a two-fan cooler coated in matte white, which immediately sets it apart from the usual dark designs you see on the market. Although Colorful has not confirmed whether it is based on NVIDIA's new GB206 or GB207 architecture, either RTX 5050 or 5060, the real point of this product is the dual M.2 SSD slots on the back of the compact PCB. Users can simply slide in drives without any extra cables, and the heatsink mounting points are cleverly positioned between the storage bays and the rear I/O bracket.

Thanks to PCIe bifurcation, the card splits a standard x16 slot into eight lanes for the GPU and four lanes for each SSD, so neither graphics nor storage performance is compromised. Given that the PCIe bandwidth is more than enough for this GPU SKU, even an x8 lane is enough for performance to stay intact. By placing SSDs near active cooling, Colorful ensures they stay cool even under heavy workloads. Official specs and pricing are not yet available, but this hybrid design is sure to catch the eye of gamers and content creators looking for a sleek, high-capacity storage build.

MSI introduced the GeForce RTX 5050 Shadow 2X OC graphics card. At the time of this writing, this is MSI's only custom RTX 5050 graphics card model. The card is a premium factory-overclocked product, and is likely to be priced above the $249 NVIDIA MSRP for the RTX 5050. It is 19.6 cm in length, and 12 cm in height, while being strictly 2 slots thick. The card offers factory-overclocked speeds of 2602 MHz boost compared to 2570 MHz reference. The RTX 5050 is cooled by an aluminium fin-stack heatsink that appears to use a single nickel-plated copper heat pipe bent in an S-shape.

The heatsink is ventilated by a pair of premium MSI TorX 5.0 fans. This fan is used by MSI in many of its premium custom-design cards, it features a partially webbed impeller that's designed to maximize axial airflow. The card draws power from a single 8-pin PCIe power connector. Display outputs include three DisplayPort 2.1b and one HDMI 2.1b. Based on the 5 nm "GB207" silicon, the RTX 5050 features 2,560 CUDA cores across 20 SM, 80 Tensor cores, 20 RT cores, and 8 GB of 20 Gbps GDDR6 memory across a 128-bit wide memory interface (320 GB/s memory bandwidth). The company didn't reveal pricing.

NVIDIA today formally launched the GeForce RTX 5050 mid-range gaming GPU for desktops and laptops. The desktop GeForce RTX 5050 graphics card starts at $249, while notebooks with RTX 5050 discrete GPUs should start at around $999. Availability of both are slated for "the second half of July 2025", although NVIDIA did not specify a date. The RTX 5050 is designed for 1080p AAA gaming with medium-thru-high settings. The card offers DLSS 4 with Multi Frame Generation, which should unlock higher performance, letting you dial up the eye-candy.

The desktop GeForce RTX 5050 debuts the "GB207" silicon, which it maxes out, enabling all 20 SM present on the chip, for 2,560 CUDA cores, 80 Tensor cores, 20 RT cores, 80 TMUs, and an unspecified ROP count. The GPU is clocked at 2.31 GHz (base) with 2.57 GHz boost. The memory, on the other hand is 8 GB in size, and uses older-generation GDDR6 memory type, across a 128-bit wide memory bus. The company didn't specify memory speed. Other features include a 130 W TGP, which makes it possible for AIC partners to build cards with 6-pin PCIe power connectors, although we expect most cards to feature 8-pin PCIe. The card comes with the latest set of NVENC and NVDEC video accelerators, and the latest display engine.

NVIDIA is rumored to have advanced the launch date of its GeForce RTX 5050 mid-range GPU to July 1, 2025. It was earlier reported that the RTX 5050 would launch by the end of July. MEGAsizeGPU, a reliable source with NVIDIA leaks, says that NVIDIA informed AICs of the advancement, and that no cards will be ready to ship on that date. We are not quite sure what NVIDIA's strategy with the RTX 5050 is. The RTX 5060 launch was eclipsed by Computex, back in May. There are no major press events planned in July. One idea would be to announce the card and have AICs make it available to purchase as soon as they can, so the company could cover some ground over the mid-Summer.

The GeForce RTX 5050 is rumored to be based on the "GB207" silicon, the company's smallest chip implementing the "Blackwell" graphics architecture. This chip is rumored to physically has 20 SM, which the RTX 5050 maxes out, for 2,560 CUDA cores, 80 Tensor cores, 20 RT cores, 80 TMUs, and an unknown number of ROPs. It's expected to offer 8 GB of GDDR6 memory (either 18 Gbps or 20 Gbps), across a 128-bit wide memory bus. NVIDIA is looking to target a sub-$250 price-point with the RTX 5050, to compete with the Intel Arc B580.Update 14:44 UTC: Turns out the RTX 5050 was announced today, with availability "in the second half of July".

NVIDIA is preparing to launch a new lower mid-range graphics card SKU in July, the GeForce RTX 5050. Positioned below the RTX 5060, the RTX 5050 possibly targets a price-point under the $250 mark, looking for a slice of the pie commanded by the Intel Arc B580. We are now learning that NVIDIA is making design choices that enable it to sell this card with an aggressive price, specifically, the choice of older generation GDDR6 memory. The card will likely feature 8 GB of GDDR6 memory across a 128-bit memory interface. At this point, we don't know the memory speeds, but if we were to hazard a guess, it could be 18 Gbps, for 288 GB/s of memory bandwidth.

The RTX 5050 is also expected to debut and max out the new "GB207" silicon, the smallest chip based on the GeForce Blackwell graphics architecture. This chip is expected to come with 20 SM, for 2,560 CUDA cores, 80 Tensor cores, 20 RT cores, 80 TMUs, and an unknown number of ROPs. The RTX 5050 is expected to be given a total graphics power (TGP) value of 130 W. It will be possible to build cards with 6-pin PCIe power connectors (75 W from connector, 75 W from the PCIe slot), although we expect single 8-pin PCIe to be the standard. The 130 W TGP will make it possible to build low-profile or compact, ITX-friendly cards.

Post-Computex 2025, another hint about NVIDIA's next rung of current-gen mobile graphics cards has emerged online. Mid-week, harukaze5719 shared an intriguing screenshot of PCI IDs, accompanied by a brief message: "2d80-2dff = GB207." Most of the captured information is common knowledge, but an unreleased model was present. The "GeForce RTX 5050 Max-Q/Mobile" entry will likely translate into real life laptop/notebook form by mid-summer.

harukaze5719's leaked list seems to confirm the future deployment of a "GB207M" GPU die—likely the smallest and least potent iteration of Team Green's "Blackwell" gaming line. According to fresh reportage, a desktop sibling is supposedly being lined up for launch in July. A late April theory suggested AMD's preparing of RDNA 4 mobile "Radeon RX 9000M" GPUs; the least powerful options—reportedly based on a Navi 44 die—could compete with NVIDIA's GeForce RTX 5060 Laptop design. Rival timelines are unclear, but manufacturing partners (Lenovo, LG, MSI, Razer etc.) seem to be prepping a plethora of Team Green-based portable gaming devices.

NVIDIA is preparing some of the final SKUs for its GeForce RTX 50 series "Blackwell" graphics cards, with the last entry being the least powerful entry-level GeForce RTX 5050 GPU. The RTX 5050 is based on GB207 SKU with 2,560 CUDA cores. Running on a 128-bit but, it carries 8 GB of GDDR6 memory, with for now unknown memory bandwidth. It carries a 130 W TDP, meaning that some improvements have been made from the previous generation RTX 4050 desktop GPU. For comparison, the last-generation RTX 4050 also had 2,560 CUDA cores, but had 6 GB of memory and 100 W TDP. Given 30% higher TDP and higher memory capacity, the Blackwell revision should give decent performance bump even with the similar CUDA core configuration. As the launch is rumored for July, we are standing by for more information about performance and price targets NVIDIA envisions.

NVIDIA's oft-leaked GeForce RTX 5050 Mobile 8 GB GPU did not make a debut appearance at last week's Computex trade show. Given the very recent launch of GeForce RTX 5060 8 GB laptops, truly cost-conscious buyers will need to wait patiently for the next wave of even cheaper portable "Blackwell" hardware. Mid-way through the month, a British e-tailer inadvertently published pre-launch prices. Lenovo's least expensive option—utilizing a GeForce RTX 5050 Mobile card—sported a (likely placeholder) tag of £1149.97 (inc. VAT). So far, a lot of pre-release information has been sourced from official websites or webstores.

Yesterday, VideoCardz shared another NDA-buster—courtesy of Acer's "Predator and Nitro Gaming Laptop GPU Power Specifications page." At the time of writing, offending items are no longer present within this list. Prior to removal, three configurable Nitro AI laptop models (16, 16S, and 18) were visible with GeForce RTX 5050 specifications (alongside RTX 5060, 5070 and 5070 Ti options). Only clock speed and TGP data points were logged by the notorious investigator (see below). Currently, TechPowerUp's GPU database entry contains speculative information—a "GB207" GPU identifier was unearthed a while ago, but some insiders reckon that Team Green's GeForce RTX 5050 Mobile design will be based on "GB206" silicon. Additionally, debates continue to rage over the selection of GDDR7 or GDDR6 VRAM. So far, all launched tiers of NVIDIA's GeForce RTX 50-series have emerged with GDDR7 memory modules.

NVIDIA's unannounced GeForce RTX 5050 Mobile graphics card has emerged in almost finalized form. According to fresh reports, Laptops Direct UK listed Lenovo's "Legion 5 15IRX10" 15.1-inch (£1349.97) and "LOQ 15IRX10" 15.6-inch gaming (£1149.97) laptop models; the "83LY000AUK" and "83JE000CUK" SKUs (respectively) seem to be equipped with Team Green's truly entry-level "Blackwell" GPU. Both product descriptions hype up the GeForce RTX 5050 8 GB GPU as offering "high-level graphics performance" and being capable of taking "your visuals to new heights." Officially, this "cheapest" mobile model is supposed to be a secret—going back to late last year, leaks revealed the entire laptop-oriented GeForce RTX 50-series product stack.

Unfortunately, the latest NDA-busting disclosure coincides with similar news regarding desktop relatives. Both lines of GeForce RTX 5060 cards are due for launch next week; expertly timed with Computex 2025 events. Given NVIDIA's apparent stealthy approach, VideoCardz posits that GeForce RTX 5050 Desktop and Mobile products could be introduced in a low-key manner. So far, the firm's manufacturing partners have inadvertently let slip very basic information. The Laptops Direct leak did not include overly-detailed specifications, but momomo_us stumbled upon another unnamed example—suggesting the selection of GDDR7 VRAM. Mid-March whispers seemed to link the GeForce RTX 5050 Mobile model's destiny to this advanced memory standard.

NVIDIA is expected to reveal its GeForce RTX 5060 Mobile and RTX 5050 Mobile GPUs later this month, but a series of leaks—going back to last summer—have already spoiled the fun. Last month, leaks pointed to Razer and MSI's preparing cheaper of "cheaper" portable gaming PCs—featuring lower end "Blackwell" Mobile hardware. VideoCardz has spent time looking for more examples—recent detective work has unearthed further evidence of an imminent launch. Yesterday's investigative article put spotlights on Razer, Lenovo and LG. Team Green's manufacturing partners have inadvertently published official web material with multiple mentioning of pre-release GeForce RTX 5060 and GeForce RTX 5050 laptop-oriented solutions. Razer China has already reacted to VideoCardz's report; their Razer Blade 16 (2025) splash page no longer lists an NDA-busting GeForce RTX 5060 Mobile option.

Similarly, LG's Taiwanese office has scrubbed "5050" from a recently published new LG gram AI notebook press release. The edited line states: "NVIDIA GeForce RTX 8 GB graphics card is only available in 16Z90TR-E.AD88C2 model." On January 31 (2025), the Lenovo PC YouTube channel uploaded an unboxing of their refreshed Lenovo IdeaPad Pro 5 16" model. The video's description let slip crucial pre-release information, regarding an upcoming discrete graphics configuration: "optional latest NVIDIA GeForce RTX 5050 graphics, with a total power output of 135 W for strong performance." VideoCardz has deduced a speculative 65 W TDP rating for Team Green's entry level "Blackwell" mobile SKU. At the time of writing, Lenovo has not edited out the offending descriptor from their Lenovo IdeaPad Pro 5 16" (2025) featurette.

NVIDIA is set to unveil its RTX 5060 series graphics cards tomorrow, according to VideoCardz information, which claims NVIDIA shared launch info with some media outlets today. The announcement will include two desktop models: the RTX 5060 and RTX 5060 Ti, confirming leaks from industry sources last week. The upcoming lineup will feature three variants: RTX 5060 Ti 16 GB, RTX 5060 Ti 8 GB, and RTX 5060. All three cards will utilize identical board designs and the same GPU, allowing manufacturers to produce visually similar Ti and non-Ti models. Power requirements are expected to range from 150-180 W. NVIDIA's RTX 5060 Ti will ship with 4608 CUDA cores, representing a modest 6% increase over the previous generation RTX 4060 Ti. The most significant improvement comes from the implementation of GDDR7 memory technology, which could deliver over 50% higher bandwidth than its predecessor if NVIDIA maintains the expected 28 Gbps memory speed across all variants.

The standard RTX 5060 will feature 3840 CUDA cores paired with 8 GB of GDDR7 memory. This configuration delivers 25% more GPU cores than its predecessor and marks an upgrade in GPU tier from AD107 (XX7) to GB206 (XX6). The smaller GB207 GPU is reportedly reserved for the upcoming RTX 5050. VideoCardz's sources indicate the RTX 5060 series will hit the market in April. Tomorrow's announcement is strategically timed as an update for the Game Developers Conference (GDC), which begins next week. All models in the series will maintain the 128-bit memory bus of their predecessors while delivering significantly improved memory bandwidth—448 GB/s compared to the previous generation's 288 GB/s for the Ti model and 272 GB/s for the standard variant. The improved bandwidth stems from the introduction of GDDR7 memory.

A curiously-specced Acer Nitro N50 (N50-656) pre-built gaming PC system was spotted and then reported online, courtesy of momomo_us's diligent eye—trained on all manner of international retail and e-tail webstores. EvoPC—a small indie outfit, located in Nancy, France—has inadvertently hinted about a potential upcoming launch of NVIDIA's unannounced GeForce RTX 5060 model. VideoCardz believes that Team Green will be unveiling the lower end of its gaming-oriented "Blackwell" GPU product stack.

An imminent announcement—reportedly marked down for this week—could include another GB203-based variant; the GeForce RTX 5060 Ti—in addition, they reckon that NVIDIA will reveal a next-gen "entry-mainstream" offering; in the shape of GeForce RTX 5050. Comprehensive details leaked onto the internet over the past weekend. EvoPC's webshop advertises the "currently unavailable" Acer Nitro N50-656 system with a steep €1589 (~$1735 USD) price tag (subject to change). Their basic rundown of internal parts lists: an Intel Core i7-14700F processor, 32 GB (2 x 16 GB) DDR5 RAM, and NVIDIA's GeForce RTX 5060 GPU with 8 GB of GDDR7 SDRAM.

NVIDIA's upcoming GeForce RTX 50 series mainstream lineup has been leaked, thanks to the well-known leaker kopite7kimi, revealing the complete specification profile of the RTX 5050, RTX 5060, and RTX 5060 Ti graphics cards. The entry-level RTX 5050 features GB207-300-A1 silicon on a PG152-SKU50 board with 2,560 CUDA cores and 8 GB of the older GDDR6 memory across a 128-bit interface at 130 W, while the mid-tier RTX 5060 utilizes GB206-250-A1 silicon on a PG152-SKU25 board housing 3,840 CUDA cores with 8 GB of the latest GDDR7 memory at 150 W. The more powerful RTX 5060 Ti implements GB206-300-A1 silicon with two board versions, PG152-SKU10/15, presumably for two memory configurations. It features 4,608 CUDA cores paired with either 8 GB or 16 GB of GDDR7 memory and a 180 W power envelope.

Most notable is the adoption of next-generation GDDR7 memory technology in both RTX 5060 variants while maintaining the 128-bit memory bus across all three models. While the entry-level RTX 5050 model utilizes GDDR6 memory, the RTX 5060 and RTX 5060 Ti are awarded with the faster GDDR7. The consistent 128-bit interface suggests NVIDIA is leveraging memory compression technologies and GDDR7's increased bandwidth to deliver performance improvements without widening the memory bus. Each tier features a slight increase in CUDA core count, a modest boost over the previous generation RTX 40 series. With the RTX 5050 targeting 1080p gaming, the RTX 5060 is positioned for high-refresh 1080p and entry-level 1440p, and the RTX 5060 Ti is likely aimed at solid 1440p performance across most titles. While we don't know the exact release date, we can expect to hear more about the availability in the coming weeks.

Possible specifications of the various NVIDIA GeForce "Blackwell" gaming GPUs were leaked to the web by Kopite7kimi, a reliable source with NVIDIA leaks. These are specs of the maxed out silicon, NVIDIA will carve out several GeForce RTX 50-series SKUs based on these chips, which could end up with lower shader counts than those shown here. We've known from older reports that there will be five chips in all, the GB202 being the largest, followed by the GB203, the GB205, the GB206, and the GB207. There is a notable absence of a successor to the AD104, GA104, and TU104, because NVIDIA is trying a slightly different way to approach the performance segment with this generation.

The GB202 is the halo segment chip that will drive the possible RTX 5090 (RTX 4090 successor). This chip is endowed with 192 streaming multiprocessors (SM), or 96 texture processing clusters (TPCs). These 96 TPCs are spread across 12 graphics processing clusters (GPCs), which each have 8 of them. Assuming that "Blackwell" has the same 256 CUDA cores per TPC that the past several generations of NVIDIA gaming GPUs have had, we end up with a total CUDA core count of 24,576. Another interesting aspect about this mega-chip is memory. The GPU implements the next-generation GDDR7 memory, and uses a mammoth 512-bit memory bus. Assuming the 28 Gbps memory speed that was being rumored for NVIDIA's "Blackwell" generation, this chip has 1,792 GB/s of memory bandwidth on tap!

Speculative NVIDIA GeForce RTX 50-series "GB20X" GPU memory interface details appeared online late last week—as disclosed by the kopite7kimi social media account. The inside information aficionado—at the time—posited that the "memory interface configuration of GB20x (Blackwell) is not much different from that of AD10x (Ada Lovelace)." It was inferred that Team Green's next flagship gaming GPU (GB202) could debut with a 384-bit memory bus—kopite7kimi had "fantasized" about a potentially monstrous 512-bit spec for the "GeForce RTX 5090." A new batch of follow-up tweets—from earlier today—rips apart last week's insights. The alleged Blackwell GPU gaming lineup includes the following SKUs: GB202, GB203, GB205, GB206, GB207.

Kopite7kimi's revised thoughts point to Team Green's flagship model possessing 192 streaming multiprocessors and a 512-bit memory bus. VideoCardz decided to interact with the reliable tipster—their queries were answered promptly: "According to kopite7kimi, there's a possibility that the second-in-line GPU, named GB203, could sport half of that core count. Now the new information is that GB203 might stick to 256-bit memory bus, which would make it half of GB202 in its entirety. What this also means is that there would be no GB20x GPU with 384-bit bus." Additional speculation has NVIDIA selecting a 192-bit bus for the GB205 SKU (AKA GeForce RTX 5070). The GeForce RTX 50-series is expected to arrive later this year—industry experts are already whispering about HPC-oriented Blackwell GPUs being unveiled at next week's GTC 2024 event. A formal gaming family announcement could arrive many months later.

The next-generation GeForce RTX 50-series graphics cards will be powered by the Blackwell graphics architecture, named after American mathematician David Blackwell. kopite7kimi, a reliable source with NVIDIA leaks revealed what the lineup of GPUs behind the series could look like. It reportedly will be led by the GB202, followed by the GB203, and then the GB205 and GB206, followed by the GB207 at the entry level. What's surprising here, is the lack of a "GB204" succeeding the AD104, GA104, TU104, and a long line of successful performance-segment GPUs by NVIDIA.

The GeForce Blackwell ASIC series begins with "GB" (GeForce Blackwell) followed by a 200-series number. The last time NVIDIA used a 200-series ASIC number for GeForce GPUs was with "Maxwell," as the GPUs ended up being built on a more advanced node, and with a few more advanced features, than what the architecture was originally conceived for. For "Blackwell," the GB202 logically succeeds the AD102, GA102, TU102, and a long line of "big chips" that have powered the company's flagship client graphics cards. The GB103 succeeds AD103, as a high SIMD count GPU with a narrower memory bus than the GB202, powering the #2 and #3 SKUs in the series. There is curiously the lack of a "GB104."