Qwen3.6-27Bcmf
infosave基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
从语言到影像,从开源到 API。读懂模型能力、比较真实成本,
为你的下一个想法,找到合适的智能引擎。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
Qwen3.8-27B (48 GatedDeltaNet linear-attention layers + 16 full-attention layers, 262k context, thinking mode) packaged as a single CMF file. It runs with `cortiq`, a Rust inference engine with no Python and no ML framework: NVIDIA, AMD and Intel GPUs through Vulkan/DX12, Apple silicon through Metal, and the CPU. Use cortiq 0.8.15 or later for fast prompt processing; prebuilt binaries for Linux, macOS and Windows are on the releases page. The engine picks the GPU and its settings itself. `--no-think` asks for a direct answer; without it the model reasons first, so raise `--max-tokens` (default 256). The 4-bit files are quantized directly from the bf16 checkpoint. Perplexity: twelve 512-token windows of wikitext-2.
译文排队中 · 当前显示来源原文
Qwen3.8-Flash-Next — a 125B hybrid mixture-of-experts model, 6B active per token (36 Gated DeltaNet + 12 sparse-attention layers, 512 experts per layer, 51B n-gram table, 262k context, thinking mode) — as CMF files, text only. It runs with `cortiq`, a Rust engine with no Python. Experts stream from the memory-mapped file as tokens need them, so the default 77 GB file runs on one GPU (measured from a 12 GB VRAM budget up to a whole 32 GB card) with the rest in system RAM. **RTX 3090 (24 GB), cortiq 0.8.15: 49.1 tok/s decode; first token after a 1,000-token prompt in 9.9 s.** RTX 5090 (cortiq 0.8.8): 81.7 tok/s, 108.6 with speculative decoding. Use cortiq 0.8.15 or later (prebuilt binaries). Without `--no-think` the model reasons first; raise `--max-tokens` (default 256). Each `cortiq run` loads the model again (10–30 s to the first token); for chat, keep it loaded with `cortiq serve`.
译文排队中 · 当前显示来源原文
Spark-X2.5-4B and Spark-X2.5-1.7B (hybrid attention: three 512-token sliding-window layers per full layer, 1M-token context, thinking mode, tool calling, 200+ languages) packaged as single CMF files. They run with `cortiq`, a Rust inference engine with no Python and no ML framework: NVIDIA, AMD and Intel GPUs through Vulkan/DX12, Apple silicon through Metal, and the CPU. Use cortiq 0.8.12 or later; prebuilt binaries for Linux, macOS and Windows are on the releases page. The engine picks the GPU and its settings itself. `--no-think` asks for a direct answer; without it the model reasons in a ` ` block first, so raise `--max-tokens` (default 256). All files are quantized from the bf16 checkpoints. Perplexity: the first 4096 tokens of wikitext-2 test with the BOS token in front; the unquantized models score 10.61 (4B) and 14.49 (1.7B; transformers in float32 gives the same 14.49). Spark is sensitive to 4-bit attention and down projections, so only gate/up are 4-bit in the compact files.
译文排队中 · 当前显示来源原文
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
GGUF, introduced by the llama.cpp team on August 21st, 2023, is a new format designed to replace the outdated GGML, which is no longer maintained by llama.cpp. GGUF boasts several improvements over GGML, such as enhanced tokenization, support for special tokens, and metadata capabilities. It is also designed with extensibility in mind. The NT-Java-1.1B GGUFs are supported by llama.cpp and are compatible with a range of third-party user interfaces and libraries. For a detailed list, please refer to the beginning of this README. **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
译文排队中 · 当前显示来源原文
**InfoVis** is a unified 9B model for slide image generation and editing, fine-tuned from FLUX.2-klein-base-9B. One model handles text-to-slide generation, template-conditioned generation, multi-image grounded generation, and instruction-based slide editing. This repository contains the InfoVis transformer in diffusers format. The text encoder (Qwen3) and the VAE are loaded from the base model. Try InfoVis in the browser in the demo Space, or run it locally with the inference script from our code repository:
译文排队中 · 当前显示来源原文
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
> **Custom fork required.** This model uses a novel architecture (MLA + MoE with identity experts + N-gram embeddings) that is not supported by upstream llama.cpp. You must build from the `longcat-flash-ngram` branch of the linked fork. LongCat-Flash-Lite is a 68.5B parameter Mixture-of-Experts language model from Meituan, with only 3–4.5B parameters activated per token. It combines three architectural innovations that make it unusually efficient: Two upstream llama.cpp PRs attempted to add this architecture: - PR #19167 (ngxson) — N-gram embedding support, blocked by base model not yet being supported - PR #19182 (ngxson) — LongCat-Flash base architecture, abandoned after maintainers deemed identity experts too complex
译文排队中 · 当前显示来源原文
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
可以调整厂商、分类或输入输出条件;新模型正持续收录。
调整月度 Token 用量,直观比较不同模型的预估 API 开支。
使用已采集且单位可比的 USD / 百万 Token 渠道报价。报价为采集时快照;不计缓存优惠、阶梯价、推理额外输出与税费,请以来源的当前报价为准。
先区分输入、输出和缓存价格,再使用同一任务和用量估算总费用。图像、音频和视频可能按张、时长或分辨率计费,不能直接与 Token 单价比较。
不一定。必须逐一核对具体模型版本的许可证、商用限制及使用政策。自己部署还需要考虑显存、推理资源和运维费用。
榜单取决于测试任务、版本与评测方法。建议用实际业务样本比较准确性、时延、稳定性和成本,再选择模型。