全站搜索
AI 模型 2119
- Nex-N2.5-mini-Sangreal-ICE↗
gbuzhf
Four ICE 1.6 tiers of **nex-agi's Nex-N2.5-mini** (35B total, 3B active, Qwen3.5-MoE architecture), quantized against **Sangreal** — a purpose-built 77-bucket calibration corpus — with an importance matrix measured on Nex itself, and shipping **Qwen-Sharp v22.5.0-arrow4.0** — our agent-tuned chat template, a heavily modified derivative of peculiar-ragdoll's Qwen-Sharp (spec below). Vision works through Nex's own projector, `mmproj-Nex-N2.5-mini-Q8_0.gguf`. Mean KLD to BF16, one session. **Size-matched** = bartowski's imatrix ladder for the same model, interpolated (log-linear) at each ICE file's exact byte size; negative = ICE closer to BF16. **Paired** = the same 64 chunks scored for both files against the same reference. Code at 16K context (`code.test.raw`, n_ctx 16384 × 8), mean KLD to BF16 vs file size; lower-left is better. Raw numbers in Table 2 below and in `measurements/2026-10-03/`.
- minimax_h3_pdmd_4nfe_comfyui↗
Iwannapose
ComfyUI-format conversion of the PDMD 4-NFE LoRA (`lora_model_0.safetensors`) — the 4-step (4 NFE) student distilled from MiniMax-H3-33B with Projected Distribution Matching Distillation, rank 128, covering attention projections and both feed-forward layers of all 50 transformer blocks + 2 token-refiner blocks. Earlier uploads (base / keys_v2 / v3 / v4 / v5) were broken conversions and have been removed: most of their keys matched nothing in the ComfyUI loader (silent no-op), and v4 missed the SwiGLU gate/value remap below. Source keys are Diffusers PEFT names (`transformer.transformer_blocks.N.attn.to_q.lora_A.weight` …); target keys follow the ComfyUI H3 layout (same scheme as the proven `minimax_h3_fl2v_turbo_8step_v1.0_768p_comfyui_bf16.safetensors`):
- SambaLingo-Hungarian-Chat-i1-GGUF↗
mradermacher
- Depth-Anything-V2-Metric-Outdoor-Small-hf↗
depth-anything
This model represents a fine-tuned version of Depth Anything V2 for outdoor metric depth estimation using the synthetic Virtual KITTI datasets. Depth Anything V2 was introduced in the paper of the same name by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in this repository. **Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
- Qwen3.5-27B-Samantha-Uncensored-GGUF↗
mradermacher
- Ternary-Bonsai-2-27B-DFlash2↗
ProCreations
Updated September 18, 2026 with the best verified continued-training checkpoint: **r2-dflash**. The original Ternary Bonsai 2 27B PQ2_0 target is unchanged. The head is available as BF16 master weights and **integer Q8_0** inference weights, not FP8. Matched testing against the previous release measured **169.12 → 170.04 decode tokens/sec (+0.54%)** on eight unseen prompts with two repeats. This is a small measured gain, not a large speedup. Both arms used the same runtime and maximum 3 draft tokens. Acceptance is accepted draft tokens divided by proposed draft tokens. Throughput is total generated tokens divided by total decode time, including reasoning tokens. The complete comparison links to the preserved raw requests, outputs and counters in the same reports directory.
- sam2_hiera_tiny.fb_r896↗
timm
`timm` SAM2 (HieraDet image encoder only) weights from https://huggingface.co/facebook/sam2-hiera-tiny
- glm-5.3-flash-exlr8-k2-k3-k4↗
kindlingai
A quantization of **GLM-5.3-Flash** in the **exlr8 v2** format, built for **NVIDIA GB10**-class unified-memory devices (DGX Spark, ASUS GX10) served on one to four boxes. Every routed and shared expert is stored at three trellis widths, K2, K3 and K4, and the serving width is chosen at load time. > **These weights cannot be used without E4M3 rounding.** A loader must round every value the trellis codebook > produces to FP8 E4M3 (round to nearest even, saturating) before applying the side vectors. The encoder searched > only over E4M3-rounded values, so the rounding is part of the format: a stock EXL3 decoder, or any decoder that > skips the rounding, reads measurably worse weights from the same bytes. GLM-5.3-Flash is a MoE model: 45 layers (0-2 dense, 3-44 MoE), 288 routed experts with top-8 routing plus one shared expert per MoE layer, an MTP layer 45 and a vision tower. This repository quantizes the routed and shared experts of layers 3-45 and the dense weights of every layer, and carries the vision tower unchanged in BF16.
- Pythia-2.8B-TLDR-Iterative-SamPO-GGUF↗
mradermacher
- Ternary-Bonsai-2-27B-mlx-2bit↗
prism-ml
Prism ML Website | Whitepaper | Demo & Examples | Discord > **8.60 GB** on disk, language model + vision tower | **98.2%** of FP16 intelligence retained | **\~47 tok/s** on an Apple M5 Max laptop - **8.60 GB** on disk: a 7.67 GB language model (down from \~54 GB FP16) plus the 0.92 GB vision tower. MLX's container stores a scale and a bias per group, so the language model costs 2.25 bits/weight where the same ternary weights take 1.75 in the GGUF PTQ1_0 packing - **98.2% of FP16 intelligence retained**: 84.78 average across 14 thinking-mode benchmarks — far above the conventional IQ2_XXS build (72.59) at less than two-thirds of its footprint, and within 0.4 points of UD-Q4_K_XL at three times the footprint - **Retains thinking, reasoning, and agentic behavior** deep in the sub-4-bit regime, where conventional low-bit representations collapse: math within half a point of full precision (96.57), coding level with the baseline (89.42), agentic tool calling at 74.92 - **End-to-end ternary language weights** across embeddings, attention projections, MLP projections, and LM head, with no high-precision escape hatches behind a low-bit label — a *true* 1.72 bits per weight as a representation, 2.25 as MLX stores it; the vision tower is bundled in this pack, unquantized - **262K-token context** on-device, kept practical by the Qwen3.8-27B hybrid-attention backbone (\~75% linear attention) - **Custom ternary hybrid-attention kernels** on Apple MLX (Python, Swift) and CUDA — packed weights are consumed directly, never expanded back to FP16 - **GGUF companion**: also available as Ternary-Bonsai-2-27B-gguf for llama.cpp (CUDA, Metal, CPU), in two packings — PTQ1_0 (5.95 GB) and PQ2_0 (7.21 GB)