Alert 🚨 🚨 🚨 IBM Quantum will retire its cloud simulators on May 15th, 2024. On BlueQubit, you can run 34 qubit cloud simulations for absolutely free!

⚛️ Quantum Hackathon: @BlueQubitIO × @DukeU 📅 Oct 24, 11:30 am ET to Oct 25, 11:30 am ET 🌍 Online from anywhere or in person at Duke 🎟️ Free, all levels 👉 bit.ly/4zrSAJZ #QubitScript #DuQuantum

✨ Magic states, 10–40x cheaper. New qLDPC factory teleports CCZ magic from tricycle codes into single-shot 4D quadcycle codes, reaching logical error rates as low as 10⁻⁹ at p = 10⁻³ 📉 📖 arxiv.org/pdf/2610.10731 #QubitScript #QuantumComputing #QEC

🏢 Enterprise quantum evaluations tend to hinge on procurement questions, not physics questions. What actually gets asked in the room: 1️⃣ #Azure Quantum — several hardware vendors under one Microsoft contract. Wins on paper for organisations already standardised on Azure. 2️⃣ #AWS Braket — the same logic on AWS, with IAM, CloudWatch and existing billing already in place. Frequently the shortest path to approval. 3️⃣ #IBM Quantum Platform — the deepest hardware roadmap and the most mature enterprise programme, tied to IBM's own devices. 4️⃣ #Classiq — circuit synthesis and IP management, aimed at teams building proprietary algorithms rather than running published ones. 5️⃣ #BlueQubit — managed simulation across statevector, MPS, and Pauli path backends, with Rigetti's 108-qubit Cepheus device self-serve, and IBM Fez and Quantinuum H2 available through partnership engagements. Per-seat team plans with pooled credits. The recurring pattern: most enterprise quantum work is still simulation, and simulation capacity is the binding constraint far more often than #QPU access. Evaluations that begin with hardware qubit counts usually end up rewritten around simulation throughput. #QubitScript #QuantumComputing #EnterpriseTech #DigitalTransformation 🔗 bluequbit.io

⚡ New on BlueQubit: Pauli Path Simulator results get a new look, all on one job page. 🔢 Observable + expectation value 📋 Cost, runtime, device 🔗 Circuit diagram ⬇️ One-click downloads Up to 192 qubits. Free on CPU. 👉 app.bluequbit.io #QubitScript #QuantumComputing

🏁 “Fastest simulator” has no single answer, because the winner changes with circuit shape. Three regimes, three different leaders: 1️⃣ Shallow, wide circuits: memory allocation dominates the runtime. GPU acceleration barely helps. A well-tuned CPU simulator like Qulacs or Qiskit Aer often wins here. 2️⃣ Deep circuits: per-gate simulation time dominates, and GPU acceleration takes over. Published benchmarks put BlueQubit GPU statevector at roughly 230× AWS Braket SV1 on a 34-qubit, 34-layer circuit, rising to about 560× at 200 layers. Depth is what moves the number. 3️⃣ Structured deep circuits: Pauli path methods change the scaling for suitable circuits, and hardware matters here too. On IBM’s 127-qubit kicked Ising circuit, BlueQubit’s GPU Pauli path backend reports speedups of up to 1,400× over the same method running on CPU. PauliPropagation.jl offers an open-source implementation of the approach. One caveat that goes hand in hand with every GPU benchmark: part of that advantage comes from allocating more GPUs as qubit count increases. Faster and cheaper are separate axes, and speedup figures rarely separate them. The broader lesson from the benchmarking literature: comparing simulators on a single circuit produces a misleading ranking. Comparing the specific circuit family in question produces a useful one. Benchmark before committing. The difference between regimes is larger than the difference between vendors. #QubitScript #QuantumSimulation #Benchmarking #GPUComputing #QuantumComputing 🔗 bluequbit.io

⚛️ Estimate what a quantum circuit measures, without a quantum computer. Duke Quantum Hackathon: run by @BlueQubitIO together with DuQuantum 2026 at @DukeU. 🎓 🔹 Around 10 hidden quantum circuits 🔹 About 50 observable expectation values 🔹 Simulations run through BlueQubit by circuit ID 🔹 Scoring based on how well estimates correlate with the exact values (signal-to-noise) 🤖 AI use is encouraged. The strongest teams will pair physical intuition with AI tools to iterate fast on simulation strategies. 📅 24 hours: Oct 24, 11:30 am ET to Oct 25, 11:30 am ET 🌍 Online for everyone 👤 In person at Duke for accepted DuQuantum participants 👥 Teams of up to 5 recommended 🎟️ Free to join; compute credits provided at kickoff 📈 All levels: from quantum enthusiasts to seasoned researchers 👉 Join the hackathon: bit.ly/4zrSAJZ #QubitScript #QuantumComputing

🤖 #Claude wrote the code. A human set the gates. New QEC paper: AI-assisted search over ~10⁸ two-block codes, all on two laptops 💻 🎯 4–15x more good codes via targeted sampling 📖 Full paper: arxiv.org/pdf/2610.13880 #QubitScript #AIforScience #QuantumErrorCorrection

💸 Simulation costs become real surprisingly fast. What the free tiers actually cover: 1️⃣ Qiskit Aer, local: free and unlimited, capped by available RAM. A 32GB laptop reaches roughly 30 qubits. Costs nothing, scales nowhere. 2️⃣ IBM Quantum Platform Open Plan: 10 minutes of real hardware time per month at no cost, with paid plans priced well above that. Good for validation, not iteration. 3️⃣ AWS Braket SV1: per-minute billing, with the AWS Free Tier covering one hour of simulation per month for the first 12 months. Predictable, and easy to overspend once parameter sweeps outgrow the free hour. 4️⃣ BlueQubit free tier: 34-qubit statevector on CPU, plus 96-qubit matrix product state and 192-qubit Pauli path backends, free under daily job caps. GPU and hardware jobs require credits, which start at $20/month and include $25 of usage. 5️⃣ Google Colab plus qsim: free GPU hours, manual setup. Free in money, expensive in time. The pattern across all of them: local is free until memory runs out; cloud is cheap until parameter sweeps begin. Budget for the sweep, not the single run. #QubitScript #QuantumComputing #Startups #QuantumSimulation #CloudComputing 🔗 bluequbit.io

Last Call: apply by October 14⏳ What a Quantum Flywheel team gets for 3 months 👇 🖥️ QPU time on the latest IBM quantum computers, the core of the program ⚡ Amazon EC2 GPU instances plus large 16- to 120-core CPU machines for simulation, circuit compilation, and verification runs 🤖 Frontier AI research and coding assistants to go from idea to pipeline in days 🧪 The BlueQubit platform, a quantum-native dev environment with GPU-accelerated simulation $150K program budget across @BlueQubitIO, @awscloud compute, @IBM quantum compute, and AI tokens. The point is simple: a good idea should not die because the compute bill is too high. 🔓 ⏳ Open call closes October 14, 2026. 👉 Apply: lnkd.in/dzHS9mqy #BlueQubit #QuantumResearch #IBMQuantum #AWS #NVIDIA #ResearchGrants

🧙 New paper: Magic Quantum Code Surgery Fault-tolerant Clifford measurements on any QLDPC code via gauging. Magic states with no post-selection. 📖 arxiv.org/pdf/2610.03920 ⚡ app.bluequbit.io #QubitScript #QuantumErrorCorrection

#New in BlueQubit: a redesigned results view for MPS simulations. Every completed MPS job now shows: - Circuit metadata at a glance: bond dimension, truncation threshold, MPS build time, queue and run time - Observables chart with expectation values for every Pauli string, so you can spot patterns across qubits instantly - A sortable table with exact values, plus one-click download for both metadata and results Run large circuits with MPS and actually see what's happening inside them. 🌐 Try it now at app.bluequbit.io #QuantumComputing #MPS #TensorNetworks #BlueQubit

Coherence time: gone. 📉 Fidelity: questionable. 🤔 Coffee: still warm. ☕🔥 #BlueQubit #QuantumComputing #QuantumMemes

🐍 Every major quantum SDK is a Python library. The differences show up in what happens around the circuit, not in the circuit itself. 1️⃣ Qiskit — the deepest standard library. Transpiler, noise models, optimization and ML modules all ship in-house. Most third-party tutorials assume it. 2️⃣ PennyLane — treats quantum circuits as differentiable functions and plugs directly into PyTorch, JAX and TensorFlow. Hard to beat for hybrid models. 3️⃣ Cirq — explicit about qubit layout and gate timing. More verbose, and deliberately so, for hardware-level work. 4️⃣ Qulacs — a C++ core with Python bindings. Very fast for local simulation, much smaller ecosystem. 5️⃣ CUDA-Q — Python and C++ together, aimed at hybrid GPU workflows and multi-GPU scaling. 6️⃣ BlueQubit SDK — not a circuit language. Existing Qiskit and Cirq circuits are submitted to remote backends by changing a single argument, which keeps the circuit code portable. Worth knowing: circuits move between these more easily than expected. OpenQASM handles most of the translation. #QubitScript #QuantumComputing #Python #QuantumSDK #Qiskit 🔗 bluequbit.io

🔬 Quantum Flywheel open call closes Oct 14. Three tracks: 1️⃣ Quantum advantage on NISQ hardware 2️⃣ Breaking advantage claims classically 3️⃣ AI for new QEC codes & decoders 3-month term, credits + QPU access. 👉 bluequbit.io/quantum-flywhee… #QuantumComputing #BlueQubit

💰 Compare plans and find your fit. Every limit, credit allotment, and capability across all five plans in one view. 🆓 Free: 50k CPU jobs/day ⭐ Standard: $20/mo, unlimited CPU jobs 🚀 Premium: $120 credits, PPS on GPU 👥 Team: shared environment, 24/7 support 👉 app.bluequbit.io

⚡ Asking for a 50-qubit simulator usually means asking the wrong question. Statevector simulation doubles memory with every added qubit. Around 34 qubits, a full statevector saturates most single-node GPU memory. At 50, no cluster on Earth stores it. So 50+ qubits is not a hardware problem. It is a method problem. What actually gets used past that line: 1️⃣ Matrix product state (tensor network) — works when entanglement stays bounded. Cost scales with bond dimension, not qubit count. Available in Qiskit Aer's MPS mode, in NVIDIA cuQuantum, and as managed CPU and GPU backends on BlueQubit reaching 96 qubits. 2️⃣ Pauli path / propagation — suited to deep, structured circuits, with accuracy governed by a truncation threshold rather than by qubit count. PauliPropagation.jl is the open-source reference implementation. BlueQubit offers managed CPU and GPU versions reaching 192 qubits and up to a billion tracked Pauli paths. 3️⃣ Tensor network contraction — AWS Braket TN1 reaches 50 qubits on circuits with favourable structure. 4️⃣ Extended stabilizer methods — near-Clifford circuits only, but effectively unbounded in qubit count where they apply. Practical rule: circuit structure determines the method, and the method determines the tool. Anyone advertising 50-qubit universal statevector simulation is describing something that does not exist. #QubitScript #QuantumSimulation #TensorNetworks #QuantumComputing 🔗 bluequbit.io

🔗 An unexpected bridge: locally decodable codes and private information retrieval, used to parallelize arbitrary quantum unitaries. Depth was never the real bottleneck. Width is. 📖 Full paper: arxiv.org/pdf/2609.40351 #QubitScript #ComplexityTheory #QuantumAlgorithms

📘 New free tutorial: BQ 201: Choosing a Quantum Circuit Simulator. Statevector vs. MPS vs. Pauli-path, compared by: 🔢 Qubit count 🔗 Entanglement and bond dimension ⚡ CPU/GPU performance Pick the right one before burning compute. 👉 app.bluequbit.io/tutorials #QuantumTutorial #QuantumComputing #BlueQubit

🧭 Quantum Computing Software in 2026: six platforms, six different jobs. No single winner here. The right choice depends entirely on the workload. 1️⃣ Qiskit (@IBM) — the largest ecosystem and the most documentation. Default choice for anyone targeting IBM hardware, and the easiest place to find help when something breaks. 2️⃣ Cirq (@Google) — fine-grained circuit control and native access to Google's qsim. Strongest fit for hardware-aware NISQ research. 3️⃣ PennyLane (@XanaduAI ) — built around automatic differentiation. The clear pick for quantum machine learning and anything that needs gradients. 4️⃣ @awscloud Braket — one interface across multiple hardware vendors, on a single bill. Makes obvious sense for teams already living inside AWS. 5️⃣ @nvidia CUDA-Q — GPU-accelerated simulation for teams that own or rent their own cluster and have the engineering capacity to run it. 6️⃣ @BlueQubitIO — managed backends spanning statevector, matrix product state, and Pauli Path Simulation, plus Rigetti hardware access, with no environment setup. Framework choice and execution backend are two separate decisions. Most teams end up mixing. #QubitScript #QuantumComputing #QuantumSoftware #Qiskit #PennyLane 🔗 bluequbit.io

📉 Fewer physical qubits per logical qubit, by design. New paper lays out systematic principles for ultra-high-rate quantum codes, with compact constructions like those at check weight 10 🎯 📖 arxiv.org/pdf/2609.30069 #QubitScript #QuantumComputing #QEC