Model · Image and text to text
Google
Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from…
Open weights
apache-2.0
31.3B parameters
262,144 tokens
transformers
Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from…
Open weights
apache-2.0
31.3B parameters
262,144 tokens
transformers
Quantized version of Ateron/Gemma-4-Writers-31B. Offline-quantized W8A16 FP8 in the float-quantized format: weights in float8e4m3fn with per-output-channel symmetric scales, activations kept in bf16/fp16. For each linear layer, every output row gets its own scale starting from amax / 448, refined by an MSE clip search over ~9 clip fractions (0.8–1.0× amax) picking the lowest-error scale per row. Weights are then quantized q = e4m3(w / scale) with round-to-nearest and saturation. This per-channel + clipping scheme gives better SNR than vLLM's online per-tensor --quantization fp8 path. Only 2D linear projection weights are quantized (attention q/k/v/o, MLP gate/up/down). Embeddings, norms…
Open weights
31.3B parameters
262,144 tokens
transformers
Quantized version of Ateron/Gemma-4-Writers-31B-V2. Offline-quantized W8A16 FP8 in the float-quantized format: weights in float8e4m3fn with per-output-channel symmetric scales, activations kept in bf16/fp16. For each linear layer, every output row gets its own scale starting from amax / 448, refined by an MSE clip search over ~9 clip fractions (0.8–1.0× amax) picking the lowest-error scale per row. Weights are then quantized q = e4m3(w / scale) with round-to-nearest and saturation. This per-channel + clipping scheme gives better SNR than vLLM's online per-tensor --quantization fp8 path. Only 2D linear projection weights are quantized (attention q/k/v/o, MLP gate/up/down). Embeddings, norms…
Open weights
31.3B parameters
262,144 tokens
transformers
HyperCLOVA X KETI-HAECHI-32B is a multimodal model derived from It was developed for two primary purposes: improving Korean cultural-heritage understanding and Korean OCR, and improving tool calling and multi-step, stateful agent execution. Alongside these goals, the model retains broad multimodal, language, and coding capabilities from the base model. heritage objects, answers questions grounded in heritage images, and reads Korean text from signs, scenes, rendered text, and public documents. - Tool calling and long-horizon task execution: selects and calls tools, carries information across multiple turns, tracks changing state, and works toward an end-to-end goal over several steps.…
Open weights
other
33.3B parameters
131,072 tokens
transformers
Weight-only quantization of built to a hard 22 GB budget with the lowest perplexity achievable inside it. This is not a uniform W4A16. Bit-width was allocated by measurement: every candidate was quantized, served by vLLM, and scored on the same held-out corpus, and the budget was spent where it bought the most. - Served by vLLM's Marlin kernels throughout — no fallback kernels. - Text + MTP speculative decoding + vision all verified. 1. The MLP, GatedDeltaNet and lmhead tensors are round-to-nearest, not AutoRound-tuned. Only qproj/kproj/vproj carry AutoRound tuning (they are inherited unchanged from an AutoRound int4 run). The measurements below were taken on round-to-nearest tensors, so…
Open weights
apache-2.0
33.3B parameters
262,144 tokens
transformers
