llama.cpp fork with TQ3_1S/4S CUDA kernels — 3.5-bit WHT quantization achieving Q4s quality at 10% smaller size. Based on RaBitQ-inspired Walsh-Hadamard transform. Enables 27B models on 16GB GPUs with 15 tok/s TG, 221 tok/s PP.
☆237Nov 5, 2026Updated this week
Alternatives and similar repositories for llama.cpp-tq3
Users that are interested in llama.cpp-tq3 are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Experimental llama.cpp fork for inference research and development☆905Updated this week
- LLM inference in C/C++☆2,446Oct 22, 2026Updated 2 weeks ago
- KVarN, KV cache precision tail, low-bit quants in llama.cpp for longer context of better precision in the same VRAM☆1,189Updated this week
- Adaptive Precision for EXpert Models: MoE-aware mixed-precision quantization☆484Oct 14, 2026Updated 3 weeks ago
- llama.cpp fork with TurboQuant WHT-rotated KV cache & weight compression + Gemma 4 MTP and Qwen 3.6 NextN speculative decoding (+30-50% t…☆431Oct 28, 2026Updated last week
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- Experimental LLM inference in C/C++☆38May 15, 2026Updated 5 months ago
- LLM speculative inference server for heterogeneous hardware & consumer GPUs☆2,979Updated this week
- Run the AEON Bench suite on your own hardware: verified HuggingFace pull → serve → benchmark (text · agentic ×3 harnesses · vision · audi…☆25Oct 6, 2026Updated last month
- KV cache compression via block-diagonal rotation. Beats TurboQuant: better PPL (6.91 vs 7.07), 28% faster decode, 5.3x faster prefill, 44…☆1,057Apr 23, 2026Updated 6 months ago
- ☆68Jun 4, 2026Updated 5 months ago
- Bigger local AI on the PC you already own. Find models, chat locally, and run supported sparse models across SSD, RAM and NVIDIA GPUs wit…☆22Nov 2, 2026Updated last week
- Structured Chain-of-Thought☆221May 16, 2026Updated 5 months ago
- Custom llama.cpp fork for running EschaLabs Escha models.☆13Sep 11, 2026Updated last month
- An experimental and feature-rich inference engine built on top of llama.cpp☆83Updated this week
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- ☆7,022Jul 20, 2026Updated 3 months ago
- Unified KV-cache compression for LLM inference: 12 Python-native methods, guarded add-on composition and routing, analytical capacity sim…☆30Oct 10, 2026Updated 3 weeks ago
- ☆16Jul 21, 2026Updated 3 months ago
- EOQ: Entropy-Optimal Quantization for LLMs. 11-41% smaller than GGUF Q4_K_M with near-FP16 perplexity.☆45Sep 4, 2026Updated 2 months ago
- Autonomous self-improving 4x DGX Spark (GB10) MoA stack + LoRA loop (DSV4F router, Qwen3.6/Omni/TwoTower/Gemma). Hermes MoA routing, ~90%…☆24Aug 17, 2026Updated 2 months ago
- ☆11Jul 10, 2026Updated 3 months ago
- VENDORIZED in lucebox-hub. Fork of llama.cpp, ggml graph for lucebox inference engine☆29Jul 8, 2026Updated 4 months ago
- Deploy concurrent Hermes Agent workers on unified-memory GPUs (GB10, DGX Spark) for maximum total tok/s. Profile-isolated, kanban-coordin…☆83Jul 18, 2026Updated 3 months ago
- Concurrency Patch to enable on DSpark - DeepSeek V4 Flash/Pro C=16 can be modified for more☆12Jun 29, 2026Updated 4 months ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- World's first TurboQuant KV cache compression for llama.cpp on AMD ROCm (RX 9070 / gfx1201)☆21Apr 28, 2026Updated 6 months ago
- ROCmFPX Family for AMD Hardware and Processors. More quants and special agent quants☆431Oct 22, 2026Updated 2 weeks ago
- An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs☆1,803Updated this week
- Native runtime, integrations, and reproducible packaging for BTL-3 and BTL-3 Compact☆19Jul 24, 2026Updated 3 months ago
- Local benchmarking UI for LLMs and AI agents☆26Apr 13, 2026Updated 6 months ago
- What actually runs on an RTX 3060 12GB, if you have any card with 12GB vram, this is for you as well. measured, with commands☆163Aug 22, 2026Updated 2 months ago
- Code for Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention☆437Updated this week
- Test LLMs on real tasks. Compare models side-by-side.☆432Aug 10, 2026Updated 2 months ago
- Control panel for VLLM, Sglang, llama.cpp, exllamav3☆1,841Oct 26, 2026Updated 2 weeks ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Distil SN97 — Competitive Model Distillation on Bittensor☆35May 20, 2026Updated 5 months ago
- DSpark-style speculative decoding draft head for Qwen3.6-27B AEON: training recipe, paired eval harness, vLLM patches. +11% serve tok/s v…☆14Jul 2, 2026Updated 4 months ago
- LLM inference in C/C++☆24Oct 21, 2026Updated 2 weeks ago
- Qwen3.8-27B EXL3 (4.00 bpw) + DFlash2 speculative decoding for ExLlamaV3, validated at 262k context on a 24 GB RTX 3090☆25Sep 20, 2026Updated last month
- Do multiple agents sharing a single asymmetrically-compressed KV pool produce quality output comparable to full-precision per-agent KV ca…☆13Apr 26, 2026Updated 6 months ago
- ReasonFlow is a novel framework designed to implement o1-like reasoning capabilities in large language models.☆19Feb 25, 2025Updated last year
- Fully uncensored, capability-enhanced abliteration of Qwen3.6-27B. NVFP4 + z-lab DFlash speculative decoding (n=12) on the unified ghcr.i…☆492Oct 6, 2026Updated last month