Cohere
Cohere's open-weight 2B multilingual automatic speech recognition model for high-accuracy audio transcription.
Running this yourself: can likely run on your own machine.
---
Quality Score
---
Arena ELO
2B
Parameters
---
Context
This measures the amount of verifiable public evidence we have, not how capable the model is. A missing field means it has not been verified yet, not that its value is zero.
12 of 22 public signals
Sign in to join the discussion
0
Downloads
0
Likes
Mar 2026
Released
4/5 signals
2/4 signals
3/5 signals
1/4 signals
2/4 signals
Gaps we are still tracking
Launches
4
Pricing
2
Benchmarks
1
Research
4
General
3
Recent launch, pricing, benchmark, and API signals linked to this model or its provider.
How is AI really affecting your job? @Cohere_Labs released the Agentic Task Ecosystem, a new dataset of 690K+ tools built for AI agents. Under a test of whether a tool could independently complete an occupational task, just 2.6% passed.
View sourceAnnouncing the Cohere Transcribe model | Cohere For AI agents: a documentation index is available at the root level at /llms.txt.
Cohere Transcribe is now available on @superwhisper 💬 Push to talk and get your transcription back almost instantly. With Superwhisper, you can use Transcribe offline, integrate it into your favorite apps, and recall specialist vocabulary. https://t.co/cCXxNODcRf
View sourceCohere Transcribe just broke 1M monthly downloads since launch, reaching 2.37 million downloads total (and climbing) 🤯 Open-source, globally-focused models are the champions of AI. Thanks for helping us take one more step into the future. https://t.co/Ld15Dskslv
View sourceIntroducing Cohere Parse 5. Our newest document parsing model with the strongest price-performance on the market. Ideal for high-volume enterprise work. Get high-quality parsing accuracy while keeping per-page prices at an industry low. https://t.co/8h3b726PGm
View sourceHow is AI really affecting your job? @Cohere_Labs released the Agentic Task Ecosystem, a new dataset of 690K+ tools built for AI agents. Under a test of whether a tool could independently complete an occupational task, just 2.6% passed.
Cohere's Chief AI Officer @jpineau1 has been named to this year's TIME 100AI list, recognizing the world's most influential people in artificial intelligence. Joelle has been shaping modern research in Canada for over two decades. We couldn't be prouder to have her on our team. https://t.co/eW1hMHKDnO
Introducing Cohere Parse 5. Our newest document parsing model with the strongest price-performance on the market. Ideal for high-volume enterprise work. Get high-quality parsing accuracy while keeping per-page prices at an industry low. https://t.co/8h3b726PGm
How do you go from jailbreaking your PlayStation to CEO of Canada's leading LLM provider? You go to the University of Toronto. "When you go to @UofT, you get raised into AI." - @aidangomez https://t.co/cVzHYy5Pp2
Supertranscribe = S-tier. Cohere is now the default local model for @superwhisper onboarding. Dictate your thoughts instantly, and keep it all on your device - no wifi required. Enjoy the Transcribe x @Nativ_AI combo, @RayFernando1337 🤝 https://t.co/wAnuYk7bfD
Cohere Transcribe is now available on @superwhisper 💬 Push to talk and get your transcription back almost instantly. With Superwhisper, you can use Transcribe offline, integrate it into your favorite apps, and recall specialist vocabulary. https://t.co/cCXxNODcRf
Cohere Transcribe just broke 1M monthly downloads since launch, reaching 2.37 million downloads total (and climbing) 🤯 Open-source, globally-focused models are the champions of AI. Thanks for helping us take one more step into the future. https://t.co/Ld15Dskslv
We introduce BLARM, a feed-forward method for video-driven 3D mesh animation. Given a monocular video and a static object mesh, BLARM predicts a temporally coherent animated mesh whose motion follows the video. Rather than relying on explicit rigs or directly regressing high-dimensional vertex motion, we represent animation using a compact set of learned, time-varying rigid motion components and time-invariant vertex-to-component skinning weights. This yields a low-dimensional deformation space without requiring skeletons, cages, skinning weights, or rig annotations. Our architecture conditions geometry-derived deformation latents on video features through factorized spatial-temporal attention, then decodes rigid transformations blended by predicted skinning weights. Trained with trajectory reconstruction, entropy regularization, and motion-aware contrastive learning, BLARM produces accurate and temporally stable animations while recovering compact, interpretable motion structure from monocular video.
Video games provide a scalable source of training data for video world models, offering diverse environments, complex interactions, and abundant in-the-wild gameplay videos. However, raw gameplay footage entangles the game world with screen-space interfaces, introducing game-specific biases and irrelevant dynamics that hinder world-model training. To address this problem, we introduce GameUI-Taxonomy and G2WEngine, a full-stack framework that formalizes gameplay UI grounding and removal. G2WEngine automatically extracts reusable UI assets from real gameplay videos and synthesizes temporally coherent UI overlays on clean footage. Using this engine, we construct Game2World, comprising 96K synthetic paired videos with precise reconstruction targets and 1,079 in-the-wild clips from 303 games for realistic evaluation. Its asset library contains 5,132 verified UI elements across 21 taxonomy categories, collected from 1,010 representative gameplay frames. Based on Game2World, we propose GameCleaner, a mask-free gameplay UI removal model that combines multimodal semantic understanding with video editing capabilities. Unlike mask-based methods, GameCleaner directly identifies and removes diverse HUD elements while preserving the underlying scene content and temporal dynamics. In a controlled pilot, world models trained on UI-free gameplay improve overall VideoReward by 6.83% over those trained on UI-overlaid data. On UI-removal evaluation, GameCleaner achieves an average AAR of 95.36 on synthetic videos, outperforming the strongest temporal mask baseline by 57.3%, and obtains the best in-the-wild AAR of 80.05 with 99.8 background preservation. These results demonstrate the scalable potential of transforming Internet gameplay videos into high-quality world-model training data. Code, dataset, and model will be available at https://github.com/Dongping-Chen/Game2World.
Reliable reward models are essential for text-to-video evaluation and alignment. However, the trade-off between evaluation accuracy and inference efficiency places high demands on the quality of training supervision. Existing approaches often rely on holistic judges with fixed rubrics or open-ended reasoning, leading to incomplete inspection, unfaithful justification, and entangled attribution. We introduce FIRM-Video, a unified checklist-driven data construction framework based on a check-before-score principle: construct dimension-specific checklists, verify each criterion against temporal visual evidence, and aggregate only verified decisions. For Instruction Following, FIRM-Video decomposes prompts into weighted atomic requirements; for World Coherence, it constructs prompt-calibrated, target-specific checks grounded in visible entities and actions; and for Perceptual Quality, it applies a generic taxonomy of visual defects. The verified criteria and scores are further transformed into natural-language analyses for end-to-end reward modeling. Subsequently, we construct FIRM-Video-90K with 88,044 dimension-specific instances from 29,348 videos, and introduce FIRM-Video-Bench with 750 point-wise human annotations across 250 videos. The Qwen3-VL-based FIRM-Video-8B achieves the best overall MAE on FIRM-Video-Bench while consistently delivering the highest VBench Total, Quality, and Semantic Scores in Best-of-8 sampling across three video generators.