{"company":{"slug":"cerebras-systems","name":"Cerebras Systems","legal_name":"Cerebras Systems Inc.","aliases":["Cerebras"],"primary_layer":"L2","primary_topic_id":"ai-asics","secondary_topics":null,"hq_country":"US","hq_city":"Sunnyvale","founded_year":2016,"is_public":false,"ticker":null,"description":"Wafer-scale Engine — single-chip alternative to multi-chip GPU clusters. Filed for IPO in 2024 (deferred). Real production deployments in scientific / pharma + select hyperscaler inference."},"profile":{"snapshot":"Wafer-scale inference cloud at ~3,000 tok/s; $1.1B Series G at $8.1B valuation","stack_layer":"L2","why_join_top_2":["Capital is in hand: $1.1B Series G at $8.1B valuation closed 2025-09-30 (signal 55dbeca3) funding concrete Oklahoma City and Minneapolis data centers (signal 0248cddc) — engineering hires are backed by deployed capex, not promises.","2026 reqs (principal packaging/signal-integrity 2026-05-06, senior physical design 2026-04-22, compiler 2026-04-02) indicate a next-gen wafer-scale bring-up; you own silicon shipping at OpenAI gpt-oss launch scale (~3,000 tok/s)."],"peer_comparison":{"substitute":{"name":"AMD","delta":"AMD MI300/MI350 is the merchant GPU path buyers default to for both training and inference; Cerebras is inference-speed-only and cannot substitute for ROCm training fleets."},"direct_competitor":{"name":"Groq","delta":"Groq's LPU also chases record inference latency, but on small deterministic chips and GroqCloud; Cerebras bets on single wafer-scale silicon plus owned US data centers."}},"self_quality_score":{"notes":"Top risk ties open-weight catalog dependence to a falsifiable Q4 2026 hyperscaler scenario.","originality":7,"evidence_density":9,"anti_generic_pass":true,"reverse_hype_present":true},"strategic_position":{"top_risk":"Inference catalog rests entirely on third-party open weights (Llama 4, Qwen3, gpt-oss); if a hyperscaler ships comparable-speed inference on these same models before Q4 2026, the speed moat erodes while the funded data-center capex is already committed.","their_bet":"That a single wafer-scale chip plus owned US data centers (signals 55dbeca3, 0248cddc) wins the latency-sensitive inference market; 2026 hires for physical design, packaging/SI and ML compilers signal a next-gen WSE tape-out.","what_they_do_best":"Delivering frontier open models at speeds GPU clouds cannot match — gpt-oss-120B at ~3,000 tok/s (signal 0182c87d) and Llama 4 Maverick at ~2,500 tok/s verified by Artificial Analysis (signal 3d3634f0)."},"top_business_signals_90d":[{"summary":"Raised a $1.1B Series G at an $8.1B valuation, co-led by Fidelity and Atreides, to fund US data center and chip-supply expansion.","signal_id":"55dbeca3-ebc6-47b0-9681-ccb495aeb071","evidence_quote":"Cerebras Systems raised $1.1 billion in a Series G round at an $8.1 billion valuation, co-led by Fidelity Management & Research Company and Atreides Management"},{"summary":"Outlined an aggressive North American build-out (Oklahoma City, Minneapolis) plus G42-linked capacity targeting tens of millions of tokens/sec.","signal_id":"0248cddc-7e40-4189-8fa8-0f4082fafc5e","evidence_quote":"scaling new domestic facilities (including Oklahoma City and Minneapolis) to push aggregate Llama inference capacity toward tens of millions of tokens per second"},{"summary":"Became a launch-day inference partner for OpenAI's open-weight gpt-oss-120B, serving it at ~3,000 tokens/second.","signal_id":"0182c87d-b2fb-4165-99a9-b90c0ddaf1aa","evidence_quote":"Cerebras made gpt-oss-120B available on its inference cloud at approximately 3,000 tokens per second, positioning itself as a high-speed launch partner"}],"top_hiring_specialties_60d":[{"job_count":1,"specialty":"physical design / place-and-route / timing closure"},{"job_count":1,"specialty":"package design & signal integrity / power delivery"},{"job_count":1,"specialty":"ML compiler / graph optimization"}]},"avg_fit_score":76.60763888888889,"fit_scores":[{"overall_score":71.5833333333333,"growth_score":88,"comp_score":null,"learning_score":85,"stability_score":55,"culture_score":62,"jobseeker_profile_id":"03038a09-5290-4bb8-9295-a14cef6419b1"},{"overall_score":81.125,"growth_score":88,"comp_score":null,"learning_score":85,"stability_score":55,"culture_score":65,"jobseeker_profile_id":"f68552cf-b42f-457b-b103-621f6e37ff6d"},{"overall_score":81.7222222222222,"growth_score":88,"comp_score":null,"learning_score":85,"stability_score":55,"culture_score":68,"jobseeker_profile_id":"96f78a9c-5631-4f0a-bdd5-94ef2ef9cd69"},{"overall_score":72,"growth_score":88,"comp_score":null,"learning_score":85,"stability_score":55,"culture_score":62,"jobseeker_profile_id":"0cc4a7bc-42f9-4bee-bcb6-932be8a68adc"}],"recent_business_signals":[{"signal_type":"funding","signal_date":"2025-09-30","signal_data":{"round":"Series G","amount_usd":1100000000,"lead_investor":"Fidelity Management & Research Company and Atreides Management (co-leads)","valuation_usd":8100000000,"other_investors":"Tiger Global, Valor Equity Partners, 1789 Capital, Altimeter, Alpha Wave Global, Benchmark"},"source":"ai_synthesis_sdk"},{"signal_type":"capacity_expansion","signal_date":"2025-09-30","signal_data":{"footprint":"Multiple new North American data centers (including Oklahoma City and Minneapolis sites) plus G42-linked international capacity","target_capacity":"Aggregate Llama inference capacity targeted in the tens of millions of tokens/second across the build-out","use_of_proceeds":"US data center and chip supply expansion funded by $1.1B Series G"},"source":"ai_synthesis_sdk"},{"signal_type":"partnership","signal_date":"2025-08-05","signal_data":{"model":"gpt-oss-120B","context":"Cerebras served OpenAI's newly released open-weight gpt-oss model as a launch-day inference partner","key_specs":"~3,000 tokens/second on Cerebras Inference","partner_name":"OpenAI"},"source":"ai_synthesis_sdk"},{"signal_type":"product_launch","signal_date":"2025-08-04","signal_data":{"model":"Qwen3-Coder-480B","pricing":"Cerebras Code Pro at $50/month; Cerebras Code Max at $200/month","key_specs":"up to ~2,000 tokens/second code generation","product_name":"Cerebras Code"},"source":"ai_synthesis_sdk"},{"signal_type":"product_launch","signal_date":"2025-07-08","signal_data":{"model":"Alibaba Qwen3-235B","pricing":"$0.60 per million input tokens / $1.20 per million output tokens","key_specs":"~1,500 tokens/second, full 131K context window","product_name":"Qwen3-235B on Cerebras Inference"},"source":"ai_synthesis_sdk"},{"signal_type":"product_launch","signal_date":"2025-05-28","signal_data":{"claim":"fastest Llama 4 inference, multiples faster than GPU-based providers","model":"Meta Llama 4 Maverick (400B)","key_specs":"~2,500 tokens/second output speed, verified by Artificial Analysis","product_name":"Cerebras Inference – Llama 4 Maverick"},"source":"ai_synthesis_sdk"}],"recent_talent_signals":[{"job_title":"Principal Package and Signal Integrity Engineer","job_level":"principal","job_specialty":["package design","signal integrity","power delivery"],"posted_date":"2026-05-06"},{"job_title":"Senior Physical Design Engineer, Wafer-Scale Engine","job_level":"senior","job_specialty":["physical design","place and route","timing closure"],"posted_date":"2026-04-22"},{"job_title":"Senior Compiler Engineer, ML Graph Optimization","job_level":"senior","job_specialty":["compiler engineering","ML graph optimization"],"posted_date":"2026-04-02"},{"job_title":"RTL Design Engineer, AI Accelerator Datapath","job_level":"mid","job_specialty":["RTL design","datapath microarchitecture"],"posted_date":"2026-03-09"},{"job_title":"Staff Design Verification Engineer","job_level":"staff","job_specialty":["design verification","UVM","formal verification"],"posted_date":"2026-02-27"}],"citation_url":"/companies/cerebras-systems","last_updated":"2026-05-14T14:33:37.941141+00:00","generated_at":"2026-05-19T12:01:58.752Z"}