TheTom/turboquant_plus

• highabout#1

  Add a clear repository description

  原因：

  复制粘贴的修复

  复制

  Research and reference implementation for TurboQuant+, a KV cache compression technique for large language models (LLMs) using PolarQuant and Walsh-Hadamard rotation, achieving 3.8-6.4x compression.

  待办完成忽略
• hightopics#2

  Add relevant topics for LLM optimization

  原因：

  复制粘贴的修复

  复制

  llm, kv-cache, transformer, quantization, compression, machine-learning, deep-learning, vllm, llama-cpp, long-context

  待办完成忽略
• highreadme#3

  Move core value proposition to the top of the README

  原因：

  当前

  # TurboQuant+
  
  > 🚀 TurboQuant KV cache compression is now in vLLM (PR #38479, merged April 2026): `--kv-cache-dtype turboquant_k8v4` and friends, with fused Triton store/decode kernels. The PR discussion drew on the asymmetric K/V findings from this repo. **Upstream llama.cpp has merged the core idea too**: Hadamard KV cache rotation (#21038, citing TurboQuant directly) with fast WHT kernels on CPU (#22631), CUDA (#23615), and Vulkan (#23687). Rotation + the stock q4_0 cache is essentially turbo4's rotation stage; the PolarQuant codebook, norm extraction, and asymmetric policies remain here and in the fork.
  
  > ### [Getting Started Guide](docs/getting-started.md) | [Configuration Recommendations](docs/turboquant-recommendations.md) | [Benchmarks](docs/benchmarks.md) | Commercial Support
  
  Implementation of TurboQuant (ICLR 2026) with implementation work, experiments, and follow-on findings beyond the base paper. Compresses transformer KV cache **3.8-6.4x** using PolarQuant + Walsh-Hadamard rotation, at near q8_0 prefill speed and ~0.9x decode throughput at long context. Validated end-to-end from 1.5B to **104B at 128K context on a MacBook** (turbo3, PPL 4.024, 74 GB peak memory).
  
  This repository is the **research home**: the Python reference implementation, the validation papers, and the benchmark data. To run TurboQuant in an inference engine, pick from the table below. Pieces that prove useful and stable get upstreamed incrementally as small, reviewable patches.

  复制粘贴的修复

  复制

  # TurboQuant+
  
  Implementation of TurboQuant (ICLR 2026) with implementation work, experiments, and follow-on findings beyond the base paper. Compresses transformer KV cache **3.8-6.4x** using PolarQuant + Walsh-Hadamard rotation, at near q8_0 prefill speed and ~0.9x decode throughput at long context. Validated end-to-end from 1.5B to **104B at 128K context on a MacBook** (turbo3, PPL 4.024, 74 GB peak memory).
  
  > 🚀 TurboQuant KV cache compression is now in vLLM (PR #38479, merged April 2026): `--kv-cache-dtype turboquant_k8v4` and friends, with fused Triton store/decode kernels. The PR discussion drew on the asymmetric K/V findings from this repo. **Upstream llama.cpp has merged the core idea too**: Hadamard KV cache rotation (#21038, citing TurboQuant directly) with fast WHT kernels on CPU (#22631), CUDA (#23615), and Vulkan (#23687). Rotation + the stock q4_0 cache is essentially turbo4's rotation stage; the PolarQuant codebook, norm extraction, and asymmetric policies remain here and in the fork.
  
  > ### [Getting Started Guide](docs/getting-started.md) | [Configuration Recommendations](docs/turboquant-recommendations.md) | [Benchmarks](docs/benchmarks.md) | Commercial Support
  
  This repository is the **research home**: the Python reference implementation, the validation papers, and the benchmark data. To run TurboQuant in an inference engine, pick from the table below. Pieces that prove useful and stable get upstreamed incrementally as small, reviewable patches.

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  How to reduce transformer KV cache memory usage for long context LLMs?

  你：未被推荐
  AI 推荐顺序：

  1. vLLM (vllm-project/vllm)
  2. DeepSpeed-MII (microsoft/DeepSpeed)
  3. Hugging Face Optimum (huggingface/optimum)
  4. Triton Inference Server (triton-inference-server/server)
  5. FasterTransformer (NVIDIA/FasterTransformer)
  6. Hugging Face Transformers (huggingface/transformers)
  7. LongFormer
  8. BigBird
  9. FlashAttention-2 (Dao-AILab/flash-attention)
  10. PyTorch (pytorch/pytorch)
  11. TensorFlow (tensorflow/tensorflow)

  AI 推荐了 11 个替代方案，却始终没点名 TheTom/turboquant_plus。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Looking for methods to efficiently run large language models on devices with limited RAM.

  你：未被推荐
  AI 推荐顺序：

  1. llama.cpp (ggerganov/llama.cpp)
  2. AWQ (mit-han-lab/awq)
  3. GPTQ (IST-DASLab/gptq)
  4. bitsandbytes (TimDettmers/bitsandbytes)
  5. Hugging Face Transformers (huggingface/transformers)
  6. accelerate (huggingface/accelerate)
  7. TinyLlama (PKU-YuanGroup/TinyLlama)
  8. LoRA (microsoft/LoRA)
  9. Hugging Face PEFT library (huggingface/peft)
  10. ONNX Runtime (microsoft/onnxruntime)
  11. OpenVINO (openvinotoolkit/openvino)
  12. TensorFlow Lite (tensorflow/tensorflow)
  13. Hugging Face Optimum (huggingface/optimum)

  AI 推荐了 13 个替代方案，却始终没点名 TheTom/turboquant_plus。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  fail

  建议：

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of TheTom/turboquant_plus?

  pass
  ⚠AI 未点名 TheTom/turboquant_plus —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts TheTom/turboquant_plus in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 TheTom/turboquant_plus

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo TheTom/turboquant_plus solve, and who is the primary audience?

  pass
  ✓AI 明确点名了 TheTom/turboquant_plus

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？