G

Groq

Hardware

AI chip company building LPUs (Language Processing Units); its own architecture page states 1,000 tokens/sec/user for inference.

Partner of NVIDIA

Total funding$3.6B

Valuation$3.5B

Employees—

Founded2016

HeadquartersMountain View, USA

StatusPrivate

Products

Selected offerings — the AI products recorded here, not the vendor's full catalogue.

• GroqCloud Model API Sold on its own

  Hosted inference service that serves open-weight models on Groq's own LPU hardware, billed per token.

Inside groqcloud

• Groq LPU (GroqChip) Hardware Included with the parent

  The Language Processing Unit (LPU) — Groq's own inference accelerator chip (GroqChip), deployed 256 per rack across Groq's own data centers. Not sold or licensed as standalone hardware today: access comes bundled into GroqCloud or into the dedicated regional deployments Groq itself builds and operates for governments and telecoms.

Closest alternatives 6

SSambaNovaPrivate · $11BSacra's SambaNova competitive analysis names Groq directly as a rival inference-cloud provider. SambaCloud and GroqCloud are both hosted-inference offerings and share the serve-a-model need.2 of 2 capabilitiesShared product typeLarger scaleSourced competitorServe a model in productionCompareCSCerebras SystemsPublic · $44.4BContrary Research's Cerebras company report, Competition section -- names Groq as Cerebras's direct rival in inference. Reciprocally corroborated by Reuters (republished on investing.com, 2025-12-24): "Groq's primary rival in the approach is Cerebras Systems."2 of 2 capabilitiesShared product typeLarger scaleSourced competitorBuild on a hosted model APIServe a model in productionCompareAAMDPublic · $785.1BSame article/quote as Arm above; also independently corroborated by Reuters (Dec 2025 Nvidia-Groq deal coverage) calling AMD a "traditional rival" challenging Nvidia "as well as startups such as Groq and Cerebras."2 of 2 capabilitiesShared product typeLarger scaleSourced competitorBuy data-centre AI computeServe a model in productionCompareIIntelPublic · $463.7BSame article/quote as Arm above.2 of 2 capabilitiesShared product typeLarger scaleSourced competitorBuy data-centre AI computeServe a model in productionCompareNNVIDIAPublic · $5.1TOverlaps on inference hardware, but the December 2025 deal cuts across the rivalry: NVIDIA agreed a $20B licence of Groq's technology and took on most of its team — a supplier relationship as much as a competing one.2 of 2 capabilitiesShared product typeLarger scaleSourced competitorSourced partnerBuy data-centre AI computeServe a model in productionCompareAAmazonPublic · $2.8TSame TechCrunch article, naming the hyperscalers as direct competitors.2 of 2 capabilitiesShared product typeLarger scaleSourced competitorBuy data-centre AI computeCompare

Similar products 28

How this list is ranked ↗

QQualcommPublic · $174.9BGroq LPU (GroqChip) and Qualcomm Cloud AI 100 Ultra do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; larger scale (public).2 of 2 capabilitiesShared product typeLarger scaleCompareAAAxelera AIPrivate · $450M raisedGroq LPU (GroqChip) and Axelera Metis AIPU do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; same layer and the same scale band (growth-stage private).2 of 2 capabilitiesShared product typeSame scale bandCompareTTenstorrentPrivate · $2.6BGroq LPU (GroqChip) and Blackhole do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; same layer and the same scale band (growth-stage private).2 of 2 capabilitiesShared product typeSame scale bandCompareHHuaweiPrivateGroq LPU (GroqChip) and Atlas do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; larger scale (private).2 of 2 capabilitiesShared product typeCompareGGraphcorePrivateGroq LPU (GroqChip) and IPU (Intelligence Processing Unit) do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; Graphcore's scale not recorded.2 of 2 capabilitiesShared product typeCompareBBroadcomPublic · $1.7TGroq LPU (GroqChip) and Broadcom custom silicon do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; larger scale (public).1 of 2 capabilitiesShared product typeLarger scaleCompareHRHorizon RoboticsPublic · $8.8BGroq LPU (GroqChip) and Journey 6 Series do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; larger scale (public).2 of 2 capabilitiesShared product typeLarger scaleCompareMTMarvell TechnologyPublic · $214.9BGroq LPU (GroqChip) and Marvell OCTEON DPU do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; larger scale (public).1 of 2 capabilitiesShared product typeLarger scaleCompareHTHailo TechnologiesAcquired ($340M deal)Groq LPU (GroqChip) and Hailo AI accelerators do comparable work on accelerator silicon; Hailo Technologies is acquired, with no independent scale.2 of 2 capabilitiesShared product typeCompareAHArm HoldingsPublic · $282.8BGroq LPU (GroqChip) and Arm CSS for Mobile 2 do comparable work on accelerator silicon; larger scale (public).2 of 2 capabilitiesShared product typeLarger scaleCompareCCiscoPublic · $433.3BGroq LPU (GroqChip) and Cisco AI PODs both serve buyers who need to buy data-centre AI compute; larger scale (public).2 of 2 capabilitiesShared product typeLarger scaleCompareDTDell TechnologiesPublic · $360.7BGroq LPU (GroqChip) and Dell Pro Max AI Desktop both serve buyers who need to buy data-centre AI compute; larger scale (public).1 of 2 capabilitiesShared product typeLarger scaleCompareAAlphabet (Google)Public · $4.2TGroq LPU (GroqChip) and Google Cloud TPU do comparable work on accelerator silicon; both also serve buyers who need to buy data-centre AI compute; larger scale (public).2 of 2 capabilitiesShared product typeLarger scaleCompareAASMLPublic · $670.1BGroq LPU (GroqChip) and ASML DUV lithography systems both serve buyers who need to buy data-centre AI compute; larger scale (public).Shared product typeLarger scaleCompareTTSMCPublic · $2TGroq LPU (GroqChip) and TSMC 3DFabric both serve buyers who need to buy data-centre AI compute; larger scale (public).Shared product typeLarger scaleCompareKKioxiaPublic · $177.9BGroq LPU (GroqChip) and KIOXIA CM9-V Series both serve buyers who need to buy data-centre AI compute; larger scale (public).Shared product typeLarger scaleCompareSESamsung ElectronicsPublic · $1.2TGroq LPU (GroqChip) and Samsung Datacenter SSD both serve buyers who need to buy data-centre AI compute; larger scale (public).Shared product typeLarger scaleCompareSHSK HynixPublic · $874.5BGroq LPU (GroqChip) and CMM-DDR5 both serve buyers who need to buy data-centre AI compute; larger scale (public).1 of 2 capabilitiesShared product typeLarger scaleCompareLLambdaPrivate · $1.5B raisedGroqCloud and Lambda Inference both serve buyers who need to build on a hosted model API and serve a model in production; similar scale (private).1 of 2 capabilitiesShared product typeCompareANArista NetworksPublic · $246.4BGroq LPU (GroqChip) and 7060X5 Series both serve buyers who need to buy data-centre AI compute; larger scale (public).Shared product typeLarger scaleCompareAMApplied MaterialsPublic · $380.7BGroq LPU (GroqChip) and Enlight 3 Optical Inspection both serve buyers who need to buy data-centre AI compute; larger scale (public).Shared product typeLarger scaleCompareFAFireworks AIPrivate · $17.5BGroqCloud and Fireworks Nexus both serve buyers who need to build on a hosted model API and serve a model in production; larger scale (private).1 of 2 capabilitiesShared product typeLarger scaleCompareRRunPodPrivateGroqCloud and Public Endpoints both serve buyers who need to build on a hosted model API and serve a model in production; RunPod's scale not recorded.1 of 2 capabilitiesShared product typeCompareCCrusoePrivate · $10BGroq LPU (GroqChip) and Crusoe Spark both serve buyers who need to buy data-centre AI compute; larger scale (private).1 of 2 capabilitiesShared product typeLarger scaleCompareOOpenRouterPrivate · $1.3BGroqCloud and OpenRouter both serve buyers who need to build on a hosted model API and serve a model in production; same layer and the same scale band (growth-stage private).1 of 2 capabilitiesShared product typeSame scale bandCompareBBasetenPrivate · $13BGroqCloud and Baseten Model APIs both serve buyers who need to build on a hosted model API and serve a model in production; larger scale (private).1 of 2 capabilitiesShared product typeLarger scaleCompareNGNebius GroupPublic · $61.6BGroqCloud and Nebius Token Factory both serve buyers who need to build on a hosted model API and serve a model in production; larger scale (public).1 of 2 capabilitiesShared product typeLarger scaleCompareOOllamaPrivate · $88M raisedGroqCloud and Ollama Cloud both serve buyers who need to build on a hosted model API and serve a model in production; similar scale (private).1 of 2 capabilitiesShared product typeCompare

Closest alternatives are pairs a vendor page itself compares. Similar products are matched from what each product does; nothing in that group is asserted as a rivalry.

Sources

Total funding

News reporting · 17 Sep 2025

Nvidia AI chip challenger Groq raises even more than expected, hits $6.9B valuation, TechCrunch, September 17, 2025. Read via stockanalysis.com's citation for this figure on 2026-08-27.

Valuation

Press release · 17 Aug 2026

Groq Closes $350 million Series A, Building the World's Leading AI Inference Cloud, Groq, 2026-08-17. post-money valuation of the round dated 2026-08-17. Supersedes the $2.8B September 2025 round: this is Groq's own announcement of a later priced round, not the NVIDIA licence fee.. Read via stockanalysis.com's citation for this figure on 2026-08-27.

Employees

Checked — none published

Checked 2026-08-27 and no citable figure found. Four sweeps were run and all came back empty for this company: (1) its own /about, /about-us, /company, /careers, /jobs and /team pages, pattern-matched for a self-published headcount; (2) stockanalysis.com's private-company page, which footnotes each figure to its source; (3) its Forbes company profile, which states a headcount and the month it is as of; (4) a targeted web search. Everything the search returned was a LinkedIn-scraping aggregator, and those disagree with each other by wide margins and cite nothing, so they are not used here. The field is null rather than carrying the round guess it held before.

Founded

Wikipedia · 27 Aug 2026

Wikipedia infobox "founded" field, read 2026-08-27.

Description

Official documentation · 9 Sep 2026

"Ultra-fast" is directionally right but Groq's own wording is "fast inference", not "ultra-fast" — narrowed to the concrete figure the architecture page actually states, "1,000 tokens/sec/user", rather than the adjective.

Open Groq in the directory

Something wrong here? Send a correction — quote this company id: groq.