正在加载视频...

视频加载失败

返回首页

MiniMax M2.7 is live on Runware on Day0! 🚀 - full end-to-end project delivery - self-evolving architecture: 30% performance gains across 100+ iterations - 97% skill adherence rate across 40+ complex tasks - 56.22% on SWE-Pro, 55.6% on VIBE-Pro, 57.0% on Terminal Bench 2

runware's profile picture

Runware

6,058 subscribers

38,628 次观看 • 6 个月前 •via X (Twitter)

科学技术财经
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

SORA 2 + SORA 2 Pro are now live on Runware ⚡️ The next evolution of generative video is here. Try it in our Playground today, or integrate it through our API with the highest available RPM. Launch links in the comments.
0:12

SORA 2 + SORA 2 Pro are now live on Runware ⚡️ The next evolution of generative video is here. Try it in our Playground today, or integrate it through our API with the highest available RPM. Launch links in the comments.

Runware

2,589,874 次观看 • 11 个月前

Grok Imagine Image Quality model is now on Runware 🎨 - strong text rendering and high realism on complex scenes - handles charts, infographics, diagrams and UI concepts well - edit existing images with text instructions
0:15

Grok Imagine Image Quality model is now on Runware 🎨 - strong text rendering and high realism on complex scenes - handles charts, infographics, diagrams and UI concepts well - edit existing images with text instructions

Runware

406,710 次观看 • 4 个月前

PixVerse Modify is live on Runware 🎨 - swap subjects & age characters - add or remove objects - change weather, lighting, or time of day - replace on-screen text - full style transfer all these changes, without touching the underlying structure of your footage.
0:19

PixVerse Modify is live on Runware 🎨 - swap subjects & age characters - add or remove objects - change weather, lighting, or time of day - replace on-screen text - full style transfer all these changes, without touching the underlying structure of your footage.

Runware

492,843 次观看 • 4 个月前

Nano Banana 2 now takes video as input 🍌 - Drop a video, get context-aware images out - Thumbnails & infographics generated from your footage - Character and style consistency across frames - Edit outputs with a reference image + prompt Feature available on Runware now
0:19

Nano Banana 2 now takes video as input 🍌 - Drop a video, get context-aware images out - Thumbnails & infographics generated from your footage - Character and style consistency across frames - Edit outputs with a reference image + prompt Feature available on Runware now

Runware

14,334 次观看 • 3 个月前

FLUX.2 is live on Runware! D0 drop! 🔥 Built by Black Forest Labs, FLUX.2 is the new SOTA model that brings better control over structure, text, and references in image generation. This is a huge step forward for image gen & image editing. And we’re here to offer you the best prices on D0. This model comes in three versions. Details below.
0:15

FLUX.2 is live on Runware! D0 drop! 🔥 Built by Black Forest Labs, FLUX.2 is the new SOTA model that brings better control over structure, text, and references in image generation. This is a huge step forward for image gen & image editing. And we’re here to offer you the best prices on D0. This model comes in three versions. Details below.

Runware

293,432 次观看 • 9 个月前

Juggernaut Z from RunDiffusion is now on Runware on D0! 🎨 - non-destructive fine-tune of Z Image Base - improved lighting, depth of field, skin texture and demographic balance - less prompting overhead to get to a usable result - compatible with existing LoRAs and Z image workflows
0:10

Juggernaut Z from RunDiffusion is now on Runware on D0! 🎨 - non-destructive fine-tune of Z Image Base - improved lighting, depth of field, skin texture and demographic balance - less prompting overhead to get to a usable result - compatible with existing LoRAs and Z image workflows

Runware

430,679 次观看 • 4 个月前

Everyone is sleeping on MiniMax's new LLM! Devs are calling it "Claude at 10% the cost" - 72.5% SWE-Multilingual. Beats Sonnet 4.5 - 88.6% VIBE-bench. Beats Gemini 3 Pro I used it to build a stock analyst that generates code, executes it & returns insights. 100% open-source!
0:19

Everyone is sleeping on MiniMax's new LLM! Devs are calling it "Claude at 10% the cost" - 72.5% SWE-Multilingual. Beats Sonnet 4.5 - 88.6% VIBE-bench. Beats Gemini 3 Pro I used it to build a stock analyst that generates code, executes it & returns insights. 100% open-source!

Avi Chawla

117,706 次观看 • 8 个月前

🚨🚨🚨 Veo 3 Fast just dropped! Now live on Runware with day 0 support. As always, best price in the industry. Starting at $0.15/sec. Volume discounts available! 💥 Give it a spin: Want $10 free to try it? 🧵👇
0:10

🚨🚨🚨 Veo 3 Fast just dropped! Now live on Runware with day 0 support. As always, best price in the industry. Starting at $0.15/sec. Volume discounts available! 💥 Give it a spin: Want $10 free to try it? 🧵👇

Runware

578,052 次观看 • 1 年前

Can a VLM see without a vision encoder? We trained one for $100, inspired by Gemma 4 12B. Latency on an M3 Pro MacBook: 112 ms -> 1.1 ms for the image path 30% lower end-to-end image+LLM The architecture is just: patchify the image -> linear projection with pos embeddings -> LLM Writeup:
0:12

Can a VLM see without a vision encoder? We trained one for $100, inspired by Gemma 4 12B. Latency on an M3 Pro MacBook: 112 ms -> 1.1 ms for the image path 30% lower end-to-end image+LLM The architecture is just: patchify the image -> linear projection with pos embeddings -> LLM Writeup:

Andi Marafioti

60,637 次观看 • 3 个月前

We are in an insane run of open-weight drops. Every modality, open source is winning. This is what an open source AI summer ☀️ looks like: 🧠 LLMs & Reasoning → DeepSeek-V4-Flash-0731 (my king 👑): 304B MoE refresh, Terminal-Bench 2.1 jumps 61.8→82.7 over the preview, DeepSWE 7.3→54.4. Closes in on Opus-4.8 on Agents' Last Exam (25.2 vs 25.7). MIT. → Muse-Glimmer-30B, from Meta (they are back!!): their first open agentic model. ~29.6B dense + perception encoder, 131k+ context, built to run fully local, no cloud. Apache 2.0. → Liquid AI LFM2.5-2.6B: 2.69B params, 131k context, 220 tok/s on an M5 Max in under 2.5GB RAM. Competitive with models 4x larger on agentic tasks. → inclusionAI Ling-3.0-flash: 124B total, only 5.1B active, ~12% the size of their old 1T flagship Ring-2.6, matches it on key benchmarks. MIT. → inclusionAI Ling-3.0-tiny: 7.9B total, 1.3B active, 86-90 tok/s on an M4 Pro MacBook at ~8GB peak memory. MIT. → NVIDIA Nemotron-3.5-Lightning-30B-A3B: hybrid Mamba-2+MoE+Attention, up to 1M context, runs on a single H100 or DGX Spark, SWE-bench Verified 52.8. → deepgrove maple-preview: 20B-A1B ternary-weight reasoner, 218 tok/s on a Mac mini M4, 5.3GB checkpoint. MIT. → BigBang-v1 (endless-frontier): fine-tuned from Qwen3.6-35B-A3B via a self-evolving generator/critic synthetic-data loop. Lands aggregate performance between DeepSeek V4 Flash (284B) and V4 Pro (1.6T), at 35B. Apache 2.0. 🎬 Video → MiniMax-H3: 33B dense omni model, native stereo audio, up to 2K/15s. 3.6k+ likes already. → Minimax-H3-Turbo (lightx2v): Apache-2.0 turbo distillation of H3 for fast inference. → Lightricks LTX-2.5: image-to-video update, custom Gemma-4-12B text encoder, a markedly stronger distilled model. 🔊 Voice → NVIDIA NemotronLabs VoiceChat-11B: full-duplex speech-to-speech, ~450ms turn-taking, #2 on open VoiceBench, and the first open full-duplex model with live tool-calling mid-conversation. 🛡️ Safety → Mistral Shieldstral-1.0-3B: 3B multimodal guardrail that takes your safety policy as plain text instead of fixed categories. Beats LlamaGuard-4-12B and ShieldGemma-9B on HarmBench (99.4) and ToxicChat (84.1) at a fraction of the size. Apache 2.0.
0:14

We are in an insane run of open-weight drops. Every modality, open source is winning. This is what an open source AI summer ☀️ looks like: 🧠 LLMs & Reasoning → DeepSeek-V4-Flash-0731 (my king 👑): 304B MoE refresh, Terminal-Bench 2.1 jumps 61.8→82.7 over the preview, DeepSWE 7.3→54.4. Closes in on Opus-4.8 on Agents' Last Exam (25.2 vs 25.7). MIT. → Muse-Glimmer-30B, from Meta (they are back!!): their first open agentic model. ~29.6B dense + perception encoder, 131k+ context, built to run fully local, no cloud. Apache 2.0. → Liquid AI LFM2.5-2.6B: 2.69B params, 131k context, 220 tok/s on an M5 Max in under 2.5GB RAM. Competitive with models 4x larger on agentic tasks. → inclusionAI Ling-3.0-flash: 124B total, only 5.1B active, ~12% the size of their old 1T flagship Ring-2.6, matches it on key benchmarks. MIT. → inclusionAI Ling-3.0-tiny: 7.9B total, 1.3B active, 86-90 tok/s on an M4 Pro MacBook at ~8GB peak memory. MIT. → NVIDIA Nemotron-3.5-Lightning-30B-A3B: hybrid Mamba-2+MoE+Attention, up to 1M context, runs on a single H100 or DGX Spark, SWE-bench Verified 52.8. → deepgrove maple-preview: 20B-A1B ternary-weight reasoner, 218 tok/s on a Mac mini M4, 5.3GB checkpoint. MIT. → BigBang-v1 (endless-frontier): fine-tuned from Qwen3.6-35B-A3B via a self-evolving generator/critic synthetic-data loop. Lands aggregate performance between DeepSeek V4 Flash (284B) and V4 Pro (1.6T), at 35B. Apache 2.0. 🎬 Video → MiniMax-H3: 33B dense omni model, native stereo audio, up to 2K/15s. 3.6k+ likes already. → Minimax-H3-Turbo (lightx2v): Apache-2.0 turbo distillation of H3 for fast inference. → Lightricks LTX-2.5: image-to-video update, custom Gemma-4-12B text encoder, a markedly stronger distilled model. 🔊 Voice → NVIDIA NemotronLabs VoiceChat-11B: full-duplex speech-to-speech, ~450ms turn-taking, #2 on open VoiceBench, and the first open full-duplex model with live tool-calling mid-conversation. 🛡️ Safety → Mistral Shieldstral-1.0-3B: 3B multimodal guardrail that takes your safety policy as plain text instead of fixed categories. Beats LlamaGuard-4-12B and ShieldGemma-9B on HarmBench (99.4) and ToxicChat (84.1) at a fraction of the size. Apache 2.0.

Victor M

55,281 次观看 • 1 个月前

MiniMax M2.5 is benchmaxed. I gave 4 models the same prompt: Create a neon "OPEN" sign in HTML. GLM 5: Clean, classic neon. Nailed it. Claude Opus 4.6: Stylized with glow. Solid. Gemini 3 Pro: Cursive with bloom lighting. Creative. MiniMax M2.5: Spelled it "O b∈ N" with broken spacing. This model ranks #2 on BridgeBench with a 59.7 score and 100% completion rate. But it can't spell a 4-letter word. Benchmarks don't tell the whole story. A model can ace structured tasks and fail basic creative prompts. High score ≠ production ready.
0:15

MiniMax M2.5 is benchmaxed. I gave 4 models the same prompt: Create a neon "OPEN" sign in HTML. GLM 5: Clean, classic neon. Nailed it. Claude Opus 4.6: Stylized with glow. Solid. Gemini 3 Pro: Cursive with bloom lighting. Creative. MiniMax M2.5: Spelled it "O b∈ N" with broken spacing. This model ranks #2 on BridgeBench with a 59.7 score and 100% completion rate. But it can't spell a 4-letter word. Benchmarks don't tell the whole story. A model can ace structured tasks and fail basic creative prompts. High score ≠ production ready.

BridgeMind

54,212 次观看 • 7 个月前

🛰️ $XPIN, the leading #DePIN project on BNB Chain, is now live on SwissBorg! XPIN Network🛰️ provides a decentralised, AI-powered communication infrastructure that delivers secure, efficient, and borderless connectivity. By combining DePIN and #PayFi technologies, XPIN has created a self-sustaining ecosystem where users can access affordable data, earn rewards, and power a revolutionary communication economy. Trade $XPIN with 15 fiat currencies or swap anything-to-anything among nearly 400 crypto assets via SwissBorg’s Meta-Exchange - finding the best available prices and deepest liquidity across 30+ CEXs & DEXs. And remember: every trade earns you $BORG cashback on your exchange fees - up to 90%, depending on your Rank! Plug into the future of connectivity. XPIN is now on SwissBorg!
0:15

🛰️ $XPIN, the leading #DePIN project on BNB Chain, is now live on SwissBorg! XPIN Network🛰️ provides a decentralised, AI-powered communication infrastructure that delivers secure, efficient, and borderless connectivity. By combining DePIN and #PayFi technologies, XPIN has created a self-sustaining ecosystem where users can access affordable data, earn rewards, and power a revolutionary communication economy. Trade $XPIN with 15 fiat currencies or swap anything-to-anything among nearly 400 crypto assets via SwissBorg’s Meta-Exchange - finding the best available prices and deepest liquidity across 30+ CEXs & DEXs. And remember: every trade earns you $BORG cashback on your exchange fees - up to 90%, depending on your Rank! Plug into the future of connectivity. XPIN is now on SwissBorg!

SwissBorg

17,309 次观看 • 10 个月前

Lightning v3.1 Pro is the most human TTS ever built. The person on the other end doesn’t care about the model. They care whether the conversation feels natural. So we fine-tuned Lightning v3.1 Pro for the details that make speech feel human: where a pause lands, which word carries the weight, and how the pace shifts when a line is a question not a statement. With 250+ expressive voices across 31+ languages, Lightning brings natural rhythm, emotion, and nuance to every interaction. Built for conversations that sound less like AI and more like someone worth listening to. Try Lightning v3.1 Pro →
0:17

Lightning v3.1 Pro is the most human TTS ever built. The person on the other end doesn’t care about the model. They care whether the conversation feels natural. So we fine-tuned Lightning v3.1 Pro for the details that make speech feel human: where a pause lands, which word carries the weight, and how the pace shifts when a line is a question not a statement. With 250+ expressive voices across 31+ languages, Lightning brings natural rhythm, emotion, and nuance to every interaction. Built for conversations that sound less like AI and more like someone worth listening to. Try Lightning v3.1 Pro →

smallest.ai

32,793 次观看 • 13 天前

SpaceXai just made grok 4.5 FREE in your coding agent starting today it's xAI's new coding model. 500k context. built for long agent sessions. no card. what you get for $0: -83.3% on terminal-bench 2.1 -64.7% on swe-bench pro - 4.2x more efficient than Opus 4.8 -500k context for big repos -$2/M in, $6/M out once it goes paid what this replaces: -Claude Opus 4.8: $15/M in, $75/M out -SuperGrok: $30-50/mo all for $0 how to set it up (2 min): >curl -fsSL | bash > grok → localhost:8000/v1 > point hermes / aider / opencode / cline to it > model: grok-4.5 Or API key at - base url Works in Hermes, Aider, OpenCode, Cline, Claude Code, and any OpenAI-compatible tool. Important: Free for a limited time. EU waits till mid-July. Rate limits apply. bookmark this before the free window closes
0:13

SpaceXai just made grok 4.5 FREE in your coding agent starting today it's xAI's new coding model. 500k context. built for long agent sessions. no card. what you get for $0: -83.3% on terminal-bench 2.1 -64.7% on swe-bench pro - 4.2x more efficient than Opus 4.8 -500k context for big repos -$2/M in, $6/M out once it goes paid what this replaces: -Claude Opus 4.8: $15/M in, $75/M out -SuperGrok: $30-50/mo all for $0 how to set it up (2 min): >curl -fsSL | bash > grok → localhost:8000/v1 > point hermes / aider / opencode / cline to it > model: grok-4.5 Or API key at - base url Works in Hermes, Aider, OpenCode, Cline, Claude Code, and any OpenAI-compatible tool. Important: Free for a limited time. EU waits till mid-July. Rate limits apply. bookmark this before the free window closes

painn

152,350 次观看 • 2 个月前

What if you kept asking an LLM to "make it better"? In some recent work at FAIR, we investigate how we can efficiently use RL to fine-tune LLMs to iteratively self-improve on their previous solutions at inference-time. Training for iterated self-improvement can be costly. The naive approach to training for K self-improvement steps leads to K times the number of rollout steps per episode. We introduce Exploratory Iteration (ExIt), an RL-based automatic curriculum method that bootstraps diverse training distributions of self-improvement tasks by upcycling the LLM's own responses at previous turns as the starting points for both self-improvement and *self-divergence.* In order to decide what task to train on next, the curriculum prioritizes sampling of partial turn histories that led to higher return variance in its GRPO group (a learnability score that comes for free). This automatic curriculum over the bootstrapped task space teaches the model how to perform iterated self-improvement while only ever training the model on single-step self-improvement tasks. We look at ExIt's impact in both single-turn (contest math problems) and multi-turn (BFCLv3 multi-turn tasks), as well as MLE-bench, where the LLM is run in a search scaffold to produce solutions to real Kaggle competitions. Across these eval settings, we find ExIt produces models with greater capacity for inference-time self-improvement compared to GRPO. Notably, ExIt models can self-improve on test tasks for many more steps than the typical solution depth encountered during training, including a 22% improvement in MLE-bench performance compared to GRPO.
0:10

What if you kept asking an LLM to "make it better"? In some recent work at FAIR, we investigate how we can efficiently use RL to fine-tune LLMs to iteratively self-improve on their previous solutions at inference-time. Training for iterated self-improvement can be costly. The naive approach to training for K self-improvement steps leads to K times the number of rollout steps per episode. We introduce Exploratory Iteration (ExIt), an RL-based automatic curriculum method that bootstraps diverse training distributions of self-improvement tasks by upcycling the LLM's own responses at previous turns as the starting points for both self-improvement and *self-divergence.* In order to decide what task to train on next, the curriculum prioritizes sampling of partial turn histories that led to higher return variance in its GRPO group (a learnability score that comes for free). This automatic curriculum over the bootstrapped task space teaches the model how to perform iterated self-improvement while only ever training the model on single-step self-improvement tasks. We look at ExIt's impact in both single-turn (contest math problems) and multi-turn (BFCLv3 multi-turn tasks), as well as MLE-bench, where the LLM is run in a search scaffold to produce solutions to real Kaggle competitions. Across these eval settings, we find ExIt produces models with greater capacity for inference-time self-improvement compared to GRPO. Notably, ExIt models can self-improve on test tasks for many more steps than the typical solution depth encountered during training, including a 22% improvement in MLE-bench performance compared to GRPO.

Minqi Jiang

41,147 次观看 • 1 年前

🚀 Sol Video Inference Engine is here! An agent-native, training-free full-stack accelerator for video diffusion. It auto-tunes cache + sparse attn + token pruning + quant + kernel fusion for any model/hardware/config. >2× end-to-end speedup on 64B Cosmos3-Super, 22B LTX-2.3 and 2B SANA-Video — near-lossless VBench quality, minimal human effort. Practical acceleration for real video gen deployment. 📄 Paper: 🌐 Project: 💻 Code: Proud of the team! 🎉
0:10

🚀 Sol Video Inference Engine is here! An agent-native, training-free full-stack accelerator for video diffusion. It auto-tunes cache + sparse attn + token pruning + quant + kernel fusion for any model/hardware/config. >2× end-to-end speedup on 64B Cosmos3-Super, 22B LTX-2.3 and 2B SANA-Video — near-lossless VBench quality, minimal human effort. Practical acceleration for real video gen deployment. 📄 Paper: 🌐 Project: 💻 Code: Proud of the team! 🎉

Enze Xie

36,653 次观看 • 2 个月前

2026 is going to be a blast 🚀 But before we ramp up for another big year, we love welcoming new and existing holders by sharing just how jam-packed 2025 was - with flawless roadmap execution across every front 💪 🔥 Asset Offerings in 2025 (100% Sold Out) CASSIA Banyan Tree - sold out in 30 hours Ramada Plaza - sold out in 11 hours Wyndham Queen - sold out in 9 hours Blossom Residences - sold out in just 2 hours 🔥 65,000+ On-chain Holders Achieved Aptos: 42k+ Base: 21k+ Ethereum: 2k+ 🔥 Product & Platform Upgrades 100% flawless product uptime Surpassed our 50th technical development sprint (100+ weeks of work) Sub-Accounts enabled (up to 10 per member) Full website revamp completed Vesting Schedule extended by 5 years Vesting schedule tracker with performance-driven unlocks added. PROPBASE referral program launched Governance module completed Mobile App Development V1 completed 🔥 Exchange & Wallet Expansion (2025) Listed on KuCoin (Top-10 CEX) Listed on XT & WEEX Bitget Wallet integration now live with Propbase 🔥 Cross-Chain & Infrastructure Developments LayerZero bridge live across Ethereum & Base Successfully migrated to native USDC across all products 🔥 Marketplace & Security Milestones Secondary Marketplace live on Propbase Nexus Two CertiK audits completed covering • Secondary Marketplace smart contracts • LayerZero integrations 🔥 Growth, Staking & Market Performance Total $PROPS staked ATH: 125M+ 29% of circulating supply staked (ATH) 24-hour trading volume ATH: $6M+ USD Total asset value tokenized: ~$1M USD $42,000 in rental yield distributed to the community 🔥 Marketing Momentum Propbase Ascend - our largest marketing campaign to date $PROPS trended on X twice #1 tweeted coin - 21 times Most tweeted RWA project across Q1, Q2, Q3 & Q4 🔥 Events & Community Engagement Token2049 Dubai: Speaker at Stake Real Estate event Token2049 Singapore: Speaker at Integra Launch event AMA with APTOS CEO Avery Ching, AMA with MEXC and 40+ AMAs, panels & interviews throughout the year Thank you everyone who have joined us throughout an amazing year of growth and determination, as RWA begins to take the main stage, the propbase team remains laser focussed on achieving our mission in becoming the largest and most recognized real estate tokenization platform in SE Asia 2025 was about execution. 2026 is about scale. 🚀💪
0:11

2026 is going to be a blast 🚀 But before we ramp up for another big year, we love welcoming new and existing holders by sharing just how jam-packed 2025 was - with flawless roadmap execution across every front 💪 🔥 Asset Offerings in 2025 (100% Sold Out) CASSIA Banyan Tree - sold out in 30 hours Ramada Plaza - sold out in 11 hours Wyndham Queen - sold out in 9 hours Blossom Residences - sold out in just 2 hours 🔥 65,000+ On-chain Holders Achieved Aptos: 42k+ Base: 21k+ Ethereum: 2k+ 🔥 Product & Platform Upgrades 100% flawless product uptime Surpassed our 50th technical development sprint (100+ weeks of work) Sub-Accounts enabled (up to 10 per member) Full website revamp completed Vesting Schedule extended by 5 years Vesting schedule tracker with performance-driven unlocks added. PROPBASE referral program launched Governance module completed Mobile App Development V1 completed 🔥 Exchange & Wallet Expansion (2025) Listed on KuCoin (Top-10 CEX) Listed on XT & WEEX Bitget Wallet integration now live with Propbase 🔥 Cross-Chain & Infrastructure Developments LayerZero bridge live across Ethereum & Base Successfully migrated to native USDC across all products 🔥 Marketplace & Security Milestones Secondary Marketplace live on Propbase Nexus Two CertiK audits completed covering • Secondary Marketplace smart contracts • LayerZero integrations 🔥 Growth, Staking & Market Performance Total $PROPS staked ATH: 125M+ 29% of circulating supply staked (ATH) 24-hour trading volume ATH: $6M+ USD Total asset value tokenized: ~$1M USD $42,000 in rental yield distributed to the community 🔥 Marketing Momentum Propbase Ascend - our largest marketing campaign to date $PROPS trended on X twice #1 tweeted coin - 21 times Most tweeted RWA project across Q1, Q2, Q3 & Q4 🔥 Events & Community Engagement Token2049 Dubai: Speaker at Stake Real Estate event Token2049 Singapore: Speaker at Integra Launch event AMA with APTOS CEO Avery Ching, AMA with MEXC and 40+ AMAs, panels & interviews throughout the year Thank you everyone who have joined us throughout an amazing year of growth and determination, as RWA begins to take the main stage, the propbase team remains laser focussed on achieving our mission in becoming the largest and most recognized real estate tokenization platform in SE Asia 2025 was about execution. 2026 is about scale. 🚀💪

Propbase

56,362 次观看 • 9 个月前

Introducing RL Environment Creator Skill Now any one can create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.
0:18

Introducing RL Environment Creator Skill Now any one can create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.

Adithya S K

46,948 次观看 • 4 个月前

This past summer we started utilizing VBT for Peak Velocity. We also used Peak Velocity in-season for 12 Weeks on Deadlift, Squat and Bench. Prior to 2025 we didn’t use any VBT and made the plunge with OVR Performance in 2025. After talks with Joe Kenn we went all in on Peak Velocity. Top Video = 2024 off season @ 90% no VBT 515# Bottom Video = 2025 In season WK 12 @ 90% PkV 565# Just by “coaches eye” we can tell the bar moves much slower on top video. Sure…We can say the athlete got stronger in 2025. But that’s doesn’t necessarily result in bar speed. Not only did he get stronger, he is moving a heavier load faster! We saw similar results across the board at the end of the season. PkV for squat, deadlift and bench….Im all in on it!
0:13

This past summer we started utilizing VBT for Peak Velocity. We also used Peak Velocity in-season for 12 Weeks on Deadlift, Squat and Bench. Prior to 2025 we didn’t use any VBT and made the plunge with OVR Performance in 2025. After talks with Joe Kenn we went all in on Peak Velocity. Top Video = 2024 off season @ 90% no VBT 515# Bottom Video = 2025 In season WK 12 @ 90% PkV 565# Just by “coaches eye” we can tell the bar moves much slower on top video. Sure…We can say the athlete got stronger in 2025. But that’s doesn’t necessarily result in bar speed. Not only did he get stronger, he is moving a heavier load faster! We saw similar results across the board at the end of the season. PkV for squat, deadlift and bench….Im all in on it!

Dr. Eric Cash

10,726 次观看 • 6 个月前