All the fundraising deals in the AI chip market (from Q2 2025 to Q2 2026)
Download our beautiful pitch about the AI chip market
In our AI chip market deck, you will find everything you need to understand the market
The AI chip market recorded 26 funding deals worth $9.94 billion between Q2 2025 and Q2 2026.
Funding accelerated sharply as AI chip startups moved from processor design to manufacturing, cloud deployment and large customer installations.
Inference accelerators attracted the most deals, while GPU companies and complete AI computing systems attracted the largest individual rounds.
And if you want to better understand this new industry, you can download our pitch covering the AI chip market.
Insights
- The AI chip market raised $9.94 billion across 26 deals, but three rounds alone represented $3.23 billion, or almost one-third of all funding.
- Q1 2026 was the strongest quarter for AI chip funding, with nine deals and $3.48 billion raised across several large processor platforms.
- AI chip funding fell by 69% between Q1 2026 and Q2 2026, although Q2 2026 still raised more than four times the Q2 2025 total.
- Inference accelerators secured 14 of the 26 deals, showing that lower-cost and lower-power model serving is the broadest opportunity in the AI chip market.
- Data-center GPU companies completed only four deals, but those transactions raised $2.65 billion because three Chinese companies used public markets.
- The average AI chip funding round was approximately $382 million, although this figure was lifted substantially by billion-dollar Cerebras and Moore Threads deals.
- Every Q1 2026 AI chip deal exceeded $50 million, showing how quickly the market moved from early development rounds toward manufacturing-scale financing.
- Groq, Cerebras Systems, Rebellions, Positron AI and Biren Technology each raised capital more than once during the five-quarter period.
- AI chip startups increasingly raised money for complete systems, including processors, racks, software, cloud capacity and customer deployment infrastructure.
- North American AI chip companies attracted the largest private rounds, while Chinese GPU companies increasingly raised capital through IPOs.
This market map, featured in our AI chip market deck, highlights top companies and startups in the AI chip market
Summary table of the funding deals in the AI chip market (last 5 quarters)
We define the AI chip market as data-center accelerators whose primary purpose is to run AI workloads, including training and inference.
We include GPUs, TPUs and other AI accelerators or ASICs sold for deployment in servers used to train or serve machine-learning models.
We exclude general-purpose CPUs, networking and memory components, and endpoint or edge chips used in phones, PCs, cars and connected devices.
You can also read our detailed analysis to understand how funding activity in the AI chip market has evolved over the last few years.
Also, you should know that we have a dedicated page, updated weekly, with all the latest fundraising deals in the AI chip market.
| Name | What they do | Amount in $ | Quarter |
|---|---|---|---|
| Lumai | Builds optical processors that use light to run AI inference in data centers. | $10M+ | Q2 2025 |
| VSORA | Develops server processors designed to run generative AI inference efficiently. | $46M | Q2 2025 |
| Biren Technology | Designs Chinese GPUs and accelerator systems for large AI training and inference workloads. | $207M | Q2 2025 |
| Arago | Develops photonic processors that run data-center AI workloads with less electricity. | $26M | Q3 2025 |
| Positron AI | Builds memory-efficient accelerator systems for serving large AI models. | $51.6M | Q3 2025 |
| FuriosaAI | Produces energy-efficient processors for large-language and multimodal model inference. | $125M | Q3 2025 |
| Groq | Provides low-latency AI inference through proprietary LPU processors and GroqCloud. | $750M | Q3 2025 |
| Rebellions | Designs energy-efficient inference processors and server systems for global data centers. | $250M | Q3 2025 |
| Cerebras Systems | Builds wafer-scale processors, AI supercomputers and cloud infrastructure. | $1.10B | Q3 2025 |
| Majestic Labs | Develops accelerator servers that can access very large pools of fast memory. | $100M+ | Q4 2025 |
| d-Matrix | Builds in-memory processors and systems for generative AI inference. | $275M | Q4 2025 |
| Moore Threads | Produces Chinese GPUs for AI training, inference, graphics and high-performance computing. | $1.13B | Q4 2025 |
| MetaX | Designs Chinese GPUs and systems for AI training, inference and advanced computing. | $596.3M | Q4 2025 |
| Biren Technology | Develops high-performance GPUs and accelerator systems for large AI workloads. | $716.85M | Q4 2025 |
| Etched | Designs transformer-specific chips for faster and more efficient language-model inference. | $500M | Q1 2026 |
| Neurophos | Develops dense photonic processors for high-throughput AI inference. | $110M | Q1 2026 |
| Positron AI | Builds low-power accelerators for memory-heavy transformer inference in data centers. | $230M | Q1 2026 |
| OLIX | Develops optical-digital processors for complex AI inference without HBM-heavy architectures. | $220M | Q1 2026 |
| Cerebras Systems | Builds wafer-scale processors, AI systems and cloud capacity for large models. | $1.00B | Q1 2026 |
| SambaNova Systems | Provides dataflow processors and complete systems for private and cloud AI deployments. | $350M | Q1 2026 |
| MatX | Develops processors built specifically for training and serving large-language models. | $500M | Q1 2026 |
| Taalas | Turns individual AI models into specialized chips with lower operating costs. | $169M | Q1 2026 |
| Rebellions | Develops inference processors and complete server infrastructure for data-center customers. | $400M | Q1 2026 |
| EVAS Intelligence | Develops RISC-V processors and platforms for AI training and inference in China. | $211M | Q2 2026 |
| Fractile | Builds memory-integrated chips and rack systems for faster AI inference. | $220M | Q2 2026 |
| Groq | Operates a global AI inference cloud powered by proprietary LPU processors. | $650M | Q2 2026 |
In our AI chip market deck, we identify pain points entrepreneurs should prioritize
How has funding activity in the AI chip market changed over time?
Q1 2026 was the most active quarter, with nine deals and $3.48 billion raised, supported by Cerebras Systems' $1 billion round and two $500 million rounds.
Q2 2025 was the least active quarter because only three AI chip deals were announced and the largest round was Biren Technology's $207 million financing.
AI chip funding fell by 69% from Q1 2026 to Q2 2026, but Q2 2026 funding was still 311% higher than in Q2 2025.
The underlying AI chip funding trend remains strong after removing the largest deals. Excluding the top transaction in each quarter, the remaining quarterly totals still ranged from $56 million to $2.48 billion.
| Quarter | Number of deals | Total raised ($) | Comment |
|---|---|---|---|
| Q2 2025 | 3 | $263M | The AI chip market remained small, with Biren Technology representing almost four-fifths of funding. |
| Q3 2025 | 6 | $2.30B | Cerebras Systems and Groq moved AI chip funding into billion-dollar quarterly territory. |
| Q4 2025 | 5 | $2.82B | Three Chinese GPU financings drove most of the quarter's AI chip funding. |
| Q1 2026 | 9 | $3.48B | A broad group of manufacturing-stage AI chip companies completed large private rounds. |
| Q2 2026 | 3 | $1.08B | Groq represented most funding, but Fractile and EVAS Intelligence also completed substantial rounds. |
| All quarters | 26 | $9.94B | The AI chip market moved rapidly from early development funding toward large-scale production capital. |
This chart, featured in our AI chip market deck, shows how Nvidia is leading in AI chips
Which startups in the AI chip market raised the largest rounds over the last months?
These startups raised the most recently in the AI chip market:
- Moore Threads raised approximately $1.13 billion through an IPO to finance new GPUs, manufacturing capacity, software development and wider deployment in China.
- Cerebras Systems raised $1.10 billion to expand wafer-scale processor production, cloud infrastructure and large customer deployments.
- Cerebras Systems raised another $1 billion to meet customer demand and strengthen its position before a possible public listing.
- Groq raised $750 million to install more LPU systems and expand its low-latency inference cloud across additional regions.
- Biren Technology raised approximately $716.85 million through an IPO to fund new GPU products, manufacturing and software compatibility.
- Groq raised $650 million to expand global inference capacity and support rising enterprise and developer demand.
- MetaX raised approximately $596.3 million through an IPO to develop advanced GPUs and expand production in China.
- Etched raised $500 million to manufacture its transformer-specific Sohu processor and complete its supporting software platform.
- MatX raised $500 million to complete its first LLM-focused processor and prepare systems for large data-center customers.
- Rebellions raised $400 million to mass-produce new inference processors, expand overseas deployments and prepare for a possible IPO.
And, yes, we do cover most of them in our our beautiful pitch about the AI chip market.
You may also want to check our ranking of the most funded startups in the AI chip market as well as our list of the most valued startups.
This chart, featured in our AI chip market deck, shows annual funding in AI chip startups
Is the AI chip market shifting toward smaller or bigger deals?
The average AI chip funding round was approximately $382 million across the five quarters covered in this analysis.
The quarterly average increased from $88 million in Q2 2025 to $564 million in Q4 2025, before falling to $360 million in Q2 2026. Large IPOs and billion-dollar private rounds caused most of these changes.
Without the largest outliers, the AI chip market still shows a shift toward bigger rounds because many companies now need capital for fabrication, system production and cloud deployment.
| Quarter | Number of deals | Average deal size | Deals below $2M | Deals above $50M |
|---|---|---|---|---|
| Q2 2025 | 3 | $87.67M | 0 | 1 |
| Q3 2025 | 6 | $383.77M | 0 | 5 |
| Q4 2025 | 5 | $563.63M | 0 | 5 |
| Q1 2026 | 9 | $386.56M | 0 | 9 |
| Q2 2026 | 3 | $360.33M | 0 | 3 |
| All quarters | 26 | $382.45M | 0 | 23 |
This chart, featured in our AI chip market deck, compares the main business model options for AI accelerator chip companies
How concentrated was funding activity in the AI chip market?
AI chip funding was highly concentrated in every quarter, although the level changed as more companies completed large rounds. The largest deal represented between 29% and 79% of quarterly funding.
Concentration was lowest in Q1 2026 because nine AI chip companies raised money, while Q2 2025 and Q2 2026 depended much more heavily on one leading transaction.
| Quarter | Number of deals | % by Top 1 | % by Top 3 | % by Top 10 |
|---|---|---|---|---|
| Q2 2025 | 3 | 78.71% | 100.00% | 100.00% |
| Q3 2025 | 6 | 47.77% | 91.20% | 100.00% |
| Q4 2025 | 5 | 40.10% | 86.69% | 100.00% |
| Q1 2026 | 9 | 28.74% | 57.49% | 100.00% |
| Q2 2026 | 3 | 60.13% | 100.00% | 100.00% |
| All quarters | 26 | 11.36% | 32.48% | 73.85% |
This chart, featured in our AI chip market deck, shows how revenue is split across customer segments in the AI chip market
Which categories in the AI chip market received the most funding?
Inference accelerators raised $3.36 billion, representing 33.8% of all AI chip funding. The category attracted capital because lower power use and faster model serving can reduce data-center operating costs.
Data-center GPUs raised $2.65 billion, or 26.7% of total funding. Three large Chinese public offerings made GPU funding much larger than the number of deals suggests.
Server AI processors raised $2.55 billion, or 25.6% of total funding. Cerebras Systems accounted for most of this amount through two large rounds for chips, systems and cloud capacity.
| Category name | Number of deals | Total raised ($) | Startups and amount |
|---|---|---|---|
| Inference Accelerators | 14 | $3.36B | Lumai ($10M+); VSORA ($46M); Arago ($26M); Positron AI ($51.6M and $230M); FuriosaAI ($125M); Groq ($750M and $650M); Rebellions ($250M and $400M); d-Matrix ($275M); Neurophos ($110M); OLIX ($220M); Fractile ($220M). |
| Data Center GPUs | 4 | $2.65B | Biren Technology ($207M and $716.85M); Moore Threads ($1.13B); MetaX ($596.3M). |
| Server AI Processors | 4 | $2.55B | Cerebras Systems ($1.10B and $1.00B); Majestic Labs ($100M+); SambaNova Systems ($350M). |
This chart, featured in our AI chip market deck, shows how AI accelerator chip technology has evolved over time
Who are the biggest investors in the AI chip market?
Atreides Management participated in four AI chip deals involving Positron AI and Cerebras Systems, making Atreides Management the most active disclosed investor.
1789 Capital participated in three rounds involving Groq and Cerebras Systems, with those rounds raising a combined $2.85 billion.
Disruptive led two Groq financings worth a combined $1.40 billion as Groq expanded its global inference infrastructure.
Fidelity participated in both Cerebras Systems rounds, which raised a combined $2.10 billion.
Arm backed Rebellions and Positron AI, two companies developing processor systems for energy-efficient AI inference.
M12 participated in rounds for d-Matrix and Neurophos, giving M12 exposure to in-memory and photonic AI processors.
Mirae Asset participated in d-Matrix and Rebellions rounds that raised a combined $675 million.
Qatar Investment Authority backed d-Matrix and Positron AI, with the two AI chip rounds raising a combined $505 million.
Disclaimer: this investor list may be incomplete; we focus on publicly disclosed lead and prominent recurring investors, so some frequent minority participants may be underrepresented. “Total funded” does not represent the amount personally invested by an individual investor. Instead, it refers to the aggregate amount raised across all fundraising rounds in which the investor participated.
| Investor | Number of deals | Total funded ($) | Startups |
|---|---|---|---|
| Atreides Management | 4 | $2.38B | Positron AI; Cerebras Systems |
| 1789 Capital | 3 | $2.85B | Groq; Cerebras Systems |
| Fidelity | 2 | $2.10B | Cerebras Systems |
| Disruptive | 2 | $1.40B | Groq |
| Mirae Asset | 2 | $675M | d-Matrix; Rebellions |
| Qatar Investment Authority | 2 | $505M | d-Matrix; Positron AI |
| Arm | 2 | $480M | Rebellions; Positron AI |
| M12 | 2 | $385M | d-Matrix; Neurophos |
This chart, featured in our AI chip market deck, shows how custom silicon demand has driven growth in the AI chip market over time
Related blog posts
- All funding deals in the AI chip market
- Which companies have raised the most funding in the AI chip market?
- How strong is fundraising in the AI chip market right now?
