Search FoxAI
AI Models 84762
- FLUX.1-Depth-dev-lora↗
Black Forest Labs
- afriberta_small↗
castorini
- gemma-4-26B-A4B-it-NVFP4↗
cklaus
NVFP4-quantized version of google/gemma-4-26B-A4B-it in **modelopt format**, optimized for vLLM on NVIDIA Blackwell (SM120) GPUs. Other available NVFP4 quantizations of this model quantize attention layers too, which causes: 1. Quality degradation from aggressive attention quantization 2. vLLM's QKV fusion scale bug (fused projections take `max()` of global scales, causing underflow) 1. Started from RedHatAI/gemma-4-26B-A4B-it-NVFP4 (compressed-tensors format) 2. Dequantized all 115 self-attention projections (q/k/v/o across 30 layers) back to BF16 using FP4-E2M1 lookup table 3. Converted from compressed-tensors to modelopt format: - `weight_packed` -> `weight` (same data) - `weight_global_scale` (divisor) -> `weight_scale_2` (1/divisor) - `input_global_scale` (divisor) -> `input_scale` (1/divisor) 4. Updated quantization config to match NVIDIA's modelopt pattern with proper `exclude_modules`
- higgs-audio-v3-stt-GGUF↗
cstr
- moonshine-tiny-GGUF↗
cstr
- Qwen3.8-Flash-Next-DERISKED-W4A16-AWQ↗
davetha
A **W4A16 (AWQ)** quantization of `Blackfrost-AI/Qwen3.8-Flash-Next-DERISKED-BF16`, an abliterated ("derisked") variant of Qwen's Qwen3.8-Flash-Next. Multimodal (image-text-to-text), MTP speculative-decode head and PLE n-gram table preserved. Built to run the 80B-class Flash-Next MoE on 2× AMD MI210 (gfx90a) at long context. - **Format:** `compressed-tensors`, **W4A16** — 4-bit weights, group size **128**, symmetric, no zero-point (loads on vLLM's `TritonW4A16` + `fused_moe_kernel_gptq_awq` fast paths). - **Method:** **AWQ** (activation-aware). Smoothing is restricted to the mathematically-exact `up_proj → down_proj` pairs. The standard `norm → q/k/v` and `norm → gate/up` smoothings are deliberately **excluded**: this architecture uses hyper-connections with a data-dependent (sigmoid) gate, so folding a smoothing scale into the norm would change the gate non-linearly and corrupt the weights. - **Quantized:** routed experts (gate/up/down) and full-attention `q/k/v/o`. - **Kept in bf16** (matching the reference recipe): the GDN linear-attention layers (recurrent state — error compounds along the sequence), the shared expert, the QSA indexer, the router gates, hyper-connections, norms, `embed_tokens`, `lm_head`, the **MTP** speculative-decode head, and the **PLE** n-gram embedding table.
- Qwen3-24B-A4B-Freedom-HQ-Thinking-Abliterated-Heretic-NEOMAX-Imatrix-GGUF↗
DavidAU
- Qwen2-1.5B-Ita↗
DeepMount00
- autojev-27b↗
denis-pplx
- Qwen.Qwen3-Reranker-8B-GGUF↗
DevQuasar
