AI Model2026
Implements MXFP4 quantization on MoE experts plus a BF16 DFlash block-diffusion drafter to propose whole-token blocks for verification, cutting memory bandwidth and backbone forward passes for trillion‑parameter text generation—targeting long‑context, agent and code workloads.
AI Model2026
Generates text from interleaved text, image, and short-video inputs using discrete diffusion and block‑autoregressive multi‑canvas sampling; built on a sparse MoE (8/128) Gemma 4 backbone and optimized for low‑latency inference and very long contexts (up to 256K tokens).
AI Model2026
Provides GGUF quantized weights and runnable instructions to run CohereLabs' North-Mini-Code-1.0 (30B A3B MoE) locally via llama.cpp or vLLM; includes quant files, build/run notes, and recommended sampling and tool-use settings for agentic coding.
Proposes a router redesign for Mixture-of-Experts (MoE) that aligns each router row with its expert's principal singular direction using Manifold Power Iteration (MPI), improving token–expert affinity. MPI applies a 'power‑then‑retract' step to push router rows toward principal singular vectors while enforcing norm constraints; the paper gives convergence theory and pretraining results on 1B–11B MoE models.
AI Model2026
A quantized 27B coder LLM fine-tuned for repository-level code generation, multi-turn tool calling, and agentic workflows — packaged for local GGUF/llama.cpp deployment with MTP speculative decoding and trace-inversion SFT. Optimized for developer tooling; experimental and not fully safety-validated.
Implements a blockwise sparse attention (MiniMax Sparse Attention) that scores and Top-k selects key-value blocks per Grouped Query Attention group to enable attention over million-token contexts. Paired with an exp-free Top-k GPU kernel and KV-outer sparse execution, it reduces per-token attention compute and yields large prefill/decoding speedups.
AI Model2026
A post-trained Mixture-of-Experts multimodal LLM with ~397B total (≈17B active) and a 1,010,000-token context for image-text-to-text and conversational tasks. Integrates SwiReasoning to switch between latent and explicit reasoning; MIT-licensed and optimized for Portuguese/English research and on-prem inference.
Applies a population-level test-time scaling strategy that uses one model as generator, verifier, refiner, and ranker to search over candidate proofs. Combines generative-verifier RL and a low false-positive verifier with tournament selection to reach competition-level performance on IMO and USAMO.
AI Model2026
An agentic multimodal coding model for long-horizon software tasks: MoE architecture (1T params, 32B activated), 256K context, image/video input, native int4 quantization and preserved chain-of-thought (thinking) mode. Tuned for multi-step coding workflows and vLLM/SGLang deployment.
AI Model2026
Provides experimental GGUF-format quantized weights for MiniMax-M3 to run local multimodal (image‑text‑video) inference via llama.cpp or Unsloth Studio. The model is very large (~428B params) and requires GPU offload or large CPU RAM; llama.cpp currently falls back from sparse to dense attention.
AI Model2026
A 3B-parameter causal LLM tuned for verifiable multi-step reasoning in math, coding and STEM using a Spectrum-to-Signal post-training pipeline (SFT, RL, offline self-distillation); not recommended for tool-calling/agent tasks.
AI Model2026
Provides a locally runnable GGUF quantized build of Kimi K2.7 Code for multimodal, coding-focused agentic workflows — a 1T-parameter MoE model with 256K context, native int4 support, preserved thinking-mode, and image/video input support.