MiniMax just open-sourced its new M2.7 coding model, clinching top scores on SWE-Pro (56.22%) and Terminal Bench 2 (57.0%). The weights are already live on Hugging Face alongside API access via the MiniMax platform, stepping up competition in open-source software engineering AI.
MiniMax just launched M2.7, a reasoning-focused model that autonomously handled 30% to 50% of its own reinforcement learning workflow, including debugging and metric analysis. It hit a 66.6% medal rate on MLE Bench Lite and scored 56.22% on SWE-Pro. Why it matters: it marks an early shift toward recursive self-improving agent architectures, available now via API.
MiniMax just introduced M2.5, a foundation model fine-tuned via large-scale reinforcement learning for real-world coding, search, and office agent tasks. It achieves an 80.2% mark on SWE-Bench Verified while pushing inference speeds up to 100 tokens per second for just $1 an hour. The model is already deployed across MiniMax Agent.
MiniMax just launched Music 2.5, an AI music model designed to solve creative control and audio realism. It introduces 14 structural tags for surgical section-by-section arrangement, a library of over 100 instruments, and lifelike vocal synthesis capturing natural breaths and vibratos.
Hailuo AI just launched Hailuo AI Agent alongside a one-week free trial for its new creative flow workspace. Users can simply chat to plan workflows and let the agent orchestrate the best models to generate images, videos, and audio on an infinite canvas.
Higgsfield just launched Draw-to-Video, letting creators turn hand-drawn sketches directly into cinematic videos instead of wrestling with text prompts. It works across models like MiniMax, Veo 3, and Seedance Pro, making visual composition much easier to control.
MiniMax just rolled out Speech 2.5, its upgraded audio model supporting voice cloning across 40 languages. It aims to wipe out robotic cadence by matching specific accents, ages, and nuanced emotional cues. If you make multilingual content, generating realistic localized voiceovers just got a whole lot simpler.
Higgsfield AI just integrated MiniMax's Hailuo video model, unlocking unlimited free generations alongside a $20,000 creator challenge on Instagram. Creators can submit short AI-generated clips until August 17 to compete for top prizes. Why it matters: It completely removes credit barriers for creators eager to test-drive frontier AI video.
MiniMax just launched MiniMax Agent, a general-purpose AI agent built to handle multi-step planning and long-horizon tasks end-to-end. It features deep tool use, multimodal capabilities, and seamless MCP support, with over 50% of the company's internal team already using it daily.
MiniMax just launched Hailuo 02, its new video generation model capable of tackling intricate movements like gymnastics. Built on a new NCR architecture, it triples model parameters and quadruples training data while boosting efficiency by 2.5x to offer native 1080p outputs. Why it matters: High-end video generation is getting physically coherent without breaking the bank.
MiniMax just open-sourced MiniMax-M1, a hybrid-attention reasoning model with a 1 million token context window and 80k-token reasoning output. Thanks to its Lightning Attention mechanism and CISPO RL algorithm, deep reasoning costs roughly 30% of DeepSeek R1's compute, with the entire RL phase trained on 512 H800s for just $535k. It rivals Gemini 2.5 Pro in long context while being completely free on the web.
MiniMax just launched Speech-02, a new text-to-speech model capable of processing up to 200,000 characters in a single prompt from any file or URL. It supports over 30 languages, unlimited voice cloning, and sub-second streaming, aiming directly at long-form audiobook and podcast generation.
MiniMax just open-sourced its new MiniMax-01 series, featuring a 456B-parameter language model and a visual multimodal model. Built on a novel Lightning Attention architecture with linear complexity, the models handle up to 4 million tokens of context with virtually zero loss. Weights are available on GitHub alongside access via MiniMax Chat.
Hailuo AI just launched I2V-01-Live, a new image-to-video model built specifically to animate 2D illustrations. It delivers smoother character motion, subtle micro-expressions, and speaking animations across various art styles, turning flat concept art into lively video.
