{"id":119990,"date":"2026-09-25T02:12:14","date_gmt":"2026-09-25T02:12:14","guid":{"rendered":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/"},"modified":"2026-09-25T02:12:14","modified_gmt":"2026-09-25T02:12:14","slug":"contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev","status":"publish","type":"post","link":"https:\/\/youzum.net\/fr\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/","title":{"rendered":"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev"},"content":{"rendered":"<p class=\"wp-block-paragraph\">Contrastive-LM has released <a href=\"https:\/\/huggingface.co\/Contrastive-LM\/CLM-v0.1-8B\">CLM-8B<\/a>, the first open model in a new class called Contrastive Language Models (CLMs). CLM does not generate text. It scores a set of candidate actions against the current state and returns probabilities. Their main baseline is <a href=\"https:\/\/en.wikipedia.org\/wiki\/Jev_(AI_model)\">Jev<\/a>, the proprietary System One model from TypeSafe AI.<\/p>\n<p class=\"wp-block-paragraph\"><strong>Is it deployable?<\/strong> Yes. The Apache-2.0 head weighs 75 MB. It runs on 1 NVIDIA GPU under Linux, with vLLM serving the Qwen3-8B encoder.<\/p>\n<h2 class=\"wp-block-heading\"><strong>What a System One Model Does<\/strong><\/h2>\n<p class=\"wp-block-paragraph\">Jev entered limited early access on 15 September 2026. It returns typed values with probabilities instead of text. CLM targets the same interface. The <a href=\"https:\/\/github.com\/Contrastive-LM\/CLM\">CLM GitHub repo<\/a> serves CLM-8B behind a TypeSafe-compatible API. It exposes 3 question types:<\/p>\n<ul class=\"wp-block-list\">\n<li><strong>Noul:<\/strong> returns the probability that a statement is true.<\/li>\n<li><strong>Choice:<\/strong> picks one option from a declared set, with probabilities.<\/li>\n<li><strong>Score:<\/strong> returns an expected level on an ordered rubric.<\/li>\n<\/ul>\n<p class=\"wp-block-paragraph\">A request written for TypeSafe\u2019s API can be replayed through CLM\u2019s Python client.<\/p>\n<h2 class=\"wp-block-heading\"><strong>How CLM Works<\/strong><\/h2>\n<p class=\"wp-block-paragraph\">CLM trains a state encoder and an action encoder with a bidirectional InfoNCE loss. Each encoder is a frozen <a href=\"https:\/\/huggingface.co\/Qwen\/Qwen3-8B\">Qwen3-8B<\/a> backbone plus a 20M-parameter trainable projection head. Training pulls each state toward the action actually taken and pushes it away from the others.<\/p>\n<p class=\"wp-block-paragraph\">At inference, CLM scores each candidate by the dot product of the state and action embeddings. A softmax over those scores becomes the answer distribution. The same primitive ranks best-of-N solutions, routes tools and answers typed decisions.<\/p>\n<p class=\"wp-block-paragraph\">This design disaggregates states and actions. In an agent loop, the state changes every step while the action set stays mostly fixed. <code>clm-serve<\/code> reserves a slab of GPU memory, similar to vLLM\u2019s KV cache, and reuses cached vectors. On 1 RTX 4090 with 3 actions, revisited states drop from 1.7 ms to 0.6 ms. The model card reports CLM running 13\u00d7 faster than Jev with about 1,000 candidates.<\/p>\n<h2 class=\"wp-block-heading\"><strong>A 3-Stage Training Recipe<\/strong><\/h2>\n<ol class=\"wp-block-list\">\n<li><strong>Pre-training<\/strong> on ~60M <a href=\"https:\/\/huggingface.co\/datasets\/Contrastive-LM\/CLM-v0.1-Pretrain-Nemotron\">Nemotron DQA question-answer pairs<\/a>.<\/li>\n<li><strong>Mid-training<\/strong> on ~30M synthetic hard negatives generated by Gemini 2.5 Flash-Lite.<\/li>\n<li><strong>Post-training<\/strong> on ~1M agent trajectories from Agent Data Protocol, Endless-Terminals and LiteCoder-Terminal-SFT.<\/li>\n<\/ol>\n<p class=\"wp-block-paragraph\">On ~100K held-out questions, pre-training alone reaches 52.1% top-1 accuracy. Mid-training lifts it to 69.2%. Training on hard negatives from the start peaks at 62.4%, then overfits.<\/p>\n<h2 class=\"wp-block-heading\"><strong>Zero-Shot Results Against Jev<\/strong><\/h2>\n<figure class=\"wp-block-table\">\n<table class=\"has-fixed-layout\">\n<thead>\n<tr>\n<th>Task<\/th>\n<th>CLM-8B latency<\/th>\n<th>Jev latency<\/th>\n<th>CLM-8B success<\/th>\n<th>Jev success<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>T-Rex game<\/td>\n<td>16.5 ms<\/td>\n<td>149.8 ms<\/td>\n<td>5\/5<\/td>\n<td>5\/5<\/td>\n<\/tr>\n<tr>\n<td>Tool calling (BFCL v4)<\/td>\n<td>76.8 ms<\/td>\n<td>125.5 ms<\/td>\n<td>95.2%<\/td>\n<td>99.2%<\/td>\n<\/tr>\n<tr>\n<td>WikiRacing<\/td>\n<td>79.8 ms<\/td>\n<td>225 ms<\/td>\n<td>26\/30<\/td>\n<td>30\/30<\/td>\n<\/tr>\n<tr>\n<td>Super Mario<\/td>\n<td>33.5 ms<\/td>\n<td>132.6 ms<\/td>\n<td>5\/5<\/td>\n<td>5\/5<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<p class=\"wp-block-paragraph\">The 9\u00d7 figure comes from the <a href=\"https:\/\/github.com\/Contrastive-LM\/CLM\/blob\/main\/examples\/t_rex\/README.md\">T-Rex game<\/a>, where actions repeat across states. CLM matches Jev on T-Rex and Super Mario. It trails on tool calling and WikiRacing while running faster on every task.<\/p>\n<h2 class=\"wp-block-heading\"><strong>CLM as a Verifier for Coding Agents<\/strong><\/h2>\n<p class=\"wp-block-paragraph\">Here a generator samples several candidate solutions and the verifier picks one. Opus 5 produced DeepSWE candidates (best-of-4). Fable 5 produced Terminal-Bench 2.1 candidates (best-of-5). The team evaluated 38 held-out DeepSWE tasks and 30 held-out Terminal-Bench 2.1 tasks. Latency was measured on an H100.<\/p>\n<figure class=\"wp-block-table\">\n<table class=\"has-fixed-layout\">\n<thead>\n<tr>\n<th>Benchmark<\/th>\n<th>Pass@1<\/th>\n<th>CLM (fine-tuned)<\/th>\n<th>Jev<\/th>\n<th>CLM latency<\/th>\n<th>Jev latency<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>DeepSWE<\/td>\n<td>73.7%<\/td>\n<td>81.6%<\/td>\n<td>71.1%<\/td>\n<td>79 ms<\/td>\n<td>449 ms<\/td>\n<\/tr>\n<tr>\n<td>Terminal-Bench 2.1<\/td>\n<td>84.0%<\/td>\n<td>87.6%<\/td>\n<td>83.1%<\/td>\n<td>32 ms<\/td>\n<td>131 ms<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<p class=\"wp-block-paragraph\">The research team reports these as new SOTA verifier results. Jev scores below pass@1 on both benchmarks, so selecting with Jev is worse than taking 1 sample. CLM runs 4.1\u00d7 to 5.7\u00d7 faster. These numbers use lightweight <a href=\"https:\/\/github.com\/Contrastive-LM\/CLM\/blob\/main\/docs\/FINETUNING.md\">fine-tuned heads<\/a>, not the zero-shot checkpoint. They are held-out subset results, not full leaderboard submissions.<\/p>\n<h2 class=\"wp-block-heading\"><strong>Interactive Explainer<\/strong><\/h2>\n<div>\n<\/div>\n<p class=\"wp-block-paragraph\">\n<h2 class=\"wp-block-heading\"><strong>Key Takeaways<\/strong><\/h2>\n<ul class=\"wp-block-list\">\n<li>CLM-8B scores candidate actions instead of generating text.<\/li>\n<li>Up to 9\u00d7 lower latency than Jev in zero-shot tests.<\/li>\n<li>Fine-tuned heads reach 81.6% on DeepSWE and 87.6% on Terminal-Bench 2.1 subsets.<\/li>\n<li>Cached state and action vectors cut agent-loop latency.<\/li>\n<li>Apache-2.0 head, self-hosted on 1 NVIDIA GPU.<\/li>\n<\/ul>\n<\/p><p class=\"wp-block-paragraph\">\n<hr class=\"wp-block-separator has-alpha-channel-opacity\" \/>\n<\/p><p class=\"wp-block-paragraph\">\n<\/p><p class=\"wp-block-paragraph\">Check out the <a href=\"https:\/\/contrastive-lm.notion.site\/\"><strong>Blog<\/strong><\/a>, <a href=\"https:\/\/github.com\/Contrastive-LM\/CLM\"><strong>Code<\/strong><\/a> and <a href=\"https:\/\/huggingface.co\/Contrastive-LM\"><strong>Data &amp; Models<\/strong><\/a>. All credit goes to the researcher of this project. Also,\u00a0feel free to follow us on\u00a0<strong><a href=\"https:\/\/x.com\/intent\/follow?screen_name=marktechpost\" target=\"_blank\" rel=\"noopener\"><mark>Twitter<\/mark><\/a><\/strong>\u00a0and don\u2019t forget to join our\u00a0<strong><a href=\"https:\/\/www.reddit.com\/r\/machinelearningnews\/\" target=\"_blank\" rel=\"noopener\">150k+ML SubReddit<\/a><\/strong>\u00a0and Subscribe to\u00a0<strong><a href=\"https:\/\/magic.beehiiv.com\/v1\/f5e63dd4-5653-4f09-83e2-321a8b1ba526?email=%7B%7Bemail%7D%7D\" target=\"_blank\" rel=\"noopener\">our Newsletter<\/a><\/strong>. Wait! are you on telegram?\u00a0<strong><a href=\"https:\/\/t.me\/machinelearningresearchnews\" target=\"_blank\" rel=\"noopener\">now you can join us on telegram as well.<\/a><\/strong><\/p>\n<p class=\"wp-block-paragraph\">Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.?\u00a0<strong><a href=\"https:\/\/forms.gle\/MJjjVDPS7whH8Ngs6\" target=\"_blank\" rel=\"noopener\"><mark>Connect with us<\/mark><\/a><\/strong><\/p>\n<p>The post <a href=\"https:\/\/www.marktechpost.com\/2026\/09\/23\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\">Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev<\/a> appeared first on <a href=\"https:\/\/www.marktechpost.com\/\">MarkTechPost<\/a>.<\/p>","protected":false},"excerpt":{"rendered":"<p>Contrastive-LM has released CLM-8B, the first open model in a new class called Contrastive Language Models (CLMs). CLM does not generate text. It scores a set of candidate actions against the current state and returns probabilities. Their main baseline is Jev, the proprietary System One model from TypeSafe AI. Is it deployable? Yes. The Apache-2.0 head weighs 75 MB. It runs on 1 NVIDIA GPU under Linux, with vLLM serving the Qwen3-8B encoder. What a System One Model Does Jev entered limited early access on 15 September 2026. It returns typed values with probabilities instead of text. CLM targets the same interface. The CLM GitHub repo serves CLM-8B behind a TypeSafe-compatible API. It exposes 3 question types: Noul: returns the probability that a statement is true. Choice: picks one option from a declared set, with probabilities. Score: returns an expected level on an ordered rubric. A request written for TypeSafe\u2019s API can be replayed through CLM\u2019s Python client. How CLM Works CLM trains a state encoder and an action encoder with a bidirectional InfoNCE loss. Each encoder is a frozen Qwen3-8B backbone plus a 20M-parameter trainable projection head. Training pulls each state toward the action actually taken and pushes it away from the others. At inference, CLM scores each candidate by the dot product of the state and action embeddings. A softmax over those scores becomes the answer distribution. The same primitive ranks best-of-N solutions, routes tools and answers typed decisions. This design disaggregates states and actions. In an agent loop, the state changes every step while the action set stays mostly fixed. clm-serve reserves a slab of GPU memory, similar to vLLM\u2019s KV cache, and reuses cached vectors. On 1 RTX 4090 with 3 actions, revisited states drop from 1.7 ms to 0.6 ms. The model card reports CLM running 13\u00d7 faster than Jev with about 1,000 candidates. A 3-Stage Training Recipe Pre-training on ~60M Nemotron DQA question-answer pairs. Mid-training on ~30M synthetic hard negatives generated by Gemini 2.5 Flash-Lite. Post-training on ~1M agent trajectories from Agent Data Protocol, Endless-Terminals and LiteCoder-Terminal-SFT. On ~100K held-out questions, pre-training alone reaches 52.1% top-1 accuracy. Mid-training lifts it to 69.2%. Training on hard negatives from the start peaks at 62.4%, then overfits. Zero-Shot Results Against Jev Task CLM-8B latency Jev latency CLM-8B success Jev success T-Rex game 16.5 ms 149.8 ms 5\/5 5\/5 Tool calling (BFCL v4) 76.8 ms 125.5 ms 95.2% 99.2% WikiRacing 79.8 ms 225 ms 26\/30 30\/30 Super Mario 33.5 ms 132.6 ms 5\/5 5\/5 The 9\u00d7 figure comes from the T-Rex game, where actions repeat across states. CLM matches Jev on T-Rex and Super Mario. It trails on tool calling and WikiRacing while running faster on every task. CLM as a Verifier for Coding Agents Here a generator samples several candidate solutions and the verifier picks one. Opus 5 produced DeepSWE candidates (best-of-4). Fable 5 produced Terminal-Bench 2.1 candidates (best-of-5). The team evaluated 38 held-out DeepSWE tasks and 30 held-out Terminal-Bench 2.1 tasks. Latency was measured on an H100. Benchmark Pass@1 CLM (fine-tuned) Jev CLM latency Jev latency DeepSWE 73.7% 81.6% 71.1% 79 ms 449 ms Terminal-Bench 2.1 84.0% 87.6% 83.1% 32 ms 131 ms The research team reports these as new SOTA verifier results. Jev scores below pass@1 on both benchmarks, so selecting with Jev is worse than taking 1 sample. CLM runs 4.1\u00d7 to 5.7\u00d7 faster. These numbers use lightweight fine-tuned heads, not the zero-shot checkpoint. They are held-out subset results, not full leaderboard submissions. Interactive Explainer Key Takeaways CLM-8B scores candidate actions instead of generating text. Up to 9\u00d7 lower latency than Jev in zero-shot tests. Fine-tuned heads reach 81.6% on DeepSWE and 87.6% on Terminal-Bench 2.1 subsets. Cached state and action vectors cut agent-loop latency. Apache-2.0 head, self-hosted on 1 NVIDIA GPU. Check out the Blog, Code and Data &amp; Models. All credit goes to the researcher of this project. Also,\u00a0feel free to follow us on\u00a0Twitter\u00a0and don\u2019t forget to join our\u00a0150k+ML SubReddit\u00a0and Subscribe to\u00a0our Newsletter. Wait! are you on telegram?\u00a0now you can join us on telegram as well. Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.?\u00a0Connect with us The post Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev appeared first on MarkTechPost.<\/p>","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"pmpro_default_level":"","site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"_pvb_checkbox_block_on_post":false,"footnotes":""},"categories":[52,5,7,1],"tags":[],"class_list":["post-119990","post","type-post","status-publish","format-standard","hentry","category-ai-club","category-committee","category-news","category-uncategorized","pmpro-has-access"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v25.3 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev - YouZum<\/title>\n<meta name=\"description\" content=\"\u0e01\u0e34\u0e08\u0e01\u0e23\u0e23\u0e21\u0e40\u0e01\u0e35\u0e48\u0e22\u0e27\u0e01\u0e31\u0e1a\u0e42\u0e14\u0e23\u0e19\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/youzum.net\/fr\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev - YouZum\" \/>\n<meta property=\"og:description\" content=\"\u0e01\u0e34\u0e08\u0e01\u0e23\u0e23\u0e21\u0e40\u0e01\u0e35\u0e48\u0e22\u0e27\u0e01\u0e31\u0e1a\u0e42\u0e14\u0e23\u0e19\" \/>\n<meta property=\"og:url\" content=\"https:\/\/youzum.net\/fr\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\" \/>\n<meta property=\"og:site_name\" content=\"YouZum\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/DroneAssociationTH\/\" \/>\n<meta property=\"article:published_time\" content=\"2026-09-25T02:12:14+00:00\" \/>\n<meta name=\"author\" content=\"admin NU\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin NU\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\"},\"author\":{\"name\":\"admin NU\",\"@id\":\"https:\/\/yousum.gpucore.co\/#\/schema\/person\/97fa48242daf3908e4d9a5f26f4a059c\"},\"headline\":\"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev\",\"datePublished\":\"2026-09-25T02:12:14+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\"},\"wordCount\":701,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/yousum.gpucore.co\/#organization\"},\"articleSection\":[\"AI\",\"Committee\",\"News\",\"Uncategorized\"],\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\",\"url\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\",\"name\":\"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev - YouZum\",\"isPartOf\":{\"@id\":\"https:\/\/yousum.gpucore.co\/#website\"},\"datePublished\":\"2026-09-25T02:12:14+00:00\",\"description\":\"\u0e01\u0e34\u0e08\u0e01\u0e23\u0e23\u0e21\u0e40\u0e01\u0e35\u0e48\u0e22\u0e27\u0e01\u0e31\u0e1a\u0e42\u0e14\u0e23\u0e19\",\"breadcrumb\":{\"@id\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/youzum.net\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/yousum.gpucore.co\/#website\",\"url\":\"https:\/\/yousum.gpucore.co\/\",\"name\":\"YouSum\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\/\/yousum.gpucore.co\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/yousum.gpucore.co\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/yousum.gpucore.co\/#organization\",\"name\":\"Drone Association Thailand\",\"url\":\"https:\/\/yousum.gpucore.co\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/yousum.gpucore.co\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/youzum.net\/wp-content\/uploads\/2024\/11\/tranparent-logo.png\",\"contentUrl\":\"https:\/\/youzum.net\/wp-content\/uploads\/2024\/11\/tranparent-logo.png\",\"width\":300,\"height\":300,\"caption\":\"Drone Association Thailand\"},\"image\":{\"@id\":\"https:\/\/yousum.gpucore.co\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/DroneAssociationTH\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/yousum.gpucore.co\/#\/schema\/person\/97fa48242daf3908e4d9a5f26f4a059c\",\"name\":\"admin NU\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/yousum.gpucore.co\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/youzum.net\/wp-content\/uploads\/avatars\/2\/1746849356-bpfull.png\",\"contentUrl\":\"https:\/\/youzum.net\/wp-content\/uploads\/avatars\/2\/1746849356-bpfull.png\",\"caption\":\"admin NU\"},\"url\":\"https:\/\/youzum.net\/fr\/members\/adminnu\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev - YouZum","description":"\u0e01\u0e34\u0e08\u0e01\u0e23\u0e23\u0e21\u0e40\u0e01\u0e35\u0e48\u0e22\u0e27\u0e01\u0e31\u0e1a\u0e42\u0e14\u0e23\u0e19","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/youzum.net\/fr\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/","og_locale":"fr_FR","og_type":"article","og_title":"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev - YouZum","og_description":"\u0e01\u0e34\u0e08\u0e01\u0e23\u0e23\u0e21\u0e40\u0e01\u0e35\u0e48\u0e22\u0e27\u0e01\u0e31\u0e1a\u0e42\u0e14\u0e23\u0e19","og_url":"https:\/\/youzum.net\/fr\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/","og_site_name":"YouZum","article_publisher":"https:\/\/www.facebook.com\/DroneAssociationTH\/","article_published_time":"2026-09-25T02:12:14+00:00","author":"admin NU","twitter_card":"summary_large_image","twitter_misc":{"\u00c9crit par":"admin NU","Dur\u00e9e de lecture estim\u00e9e":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#article","isPartOf":{"@id":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/"},"author":{"name":"admin NU","@id":"https:\/\/yousum.gpucore.co\/#\/schema\/person\/97fa48242daf3908e4d9a5f26f4a059c"},"headline":"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev","datePublished":"2026-09-25T02:12:14+00:00","mainEntityOfPage":{"@id":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/"},"wordCount":701,"commentCount":0,"publisher":{"@id":"https:\/\/yousum.gpucore.co\/#organization"},"articleSection":["AI","Committee","News","Uncategorized"],"inLanguage":"fr-FR","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/","url":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/","name":"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev - YouZum","isPartOf":{"@id":"https:\/\/yousum.gpucore.co\/#website"},"datePublished":"2026-09-25T02:12:14+00:00","description":"\u0e01\u0e34\u0e08\u0e01\u0e23\u0e23\u0e21\u0e40\u0e01\u0e35\u0e48\u0e22\u0e27\u0e01\u0e31\u0e1a\u0e42\u0e14\u0e23\u0e19","breadcrumb":{"@id":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/youzum.net\/contrastive-lm-releases-clm-8b-an-open-system-one-model-that-scores-agent-actions-up-to-9x-faster-than-jev\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/youzum.net\/"},{"@type":"ListItem","position":2,"name":"Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9\u00d7 Faster Than Jev"}]},{"@type":"WebSite","@id":"https:\/\/yousum.gpucore.co\/#website","url":"https:\/\/yousum.gpucore.co\/","name":"YouSum","description":"","publisher":{"@id":"https:\/\/yousum.gpucore.co\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/yousum.gpucore.co\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"},{"@type":"Organization","@id":"https:\/\/yousum.gpucore.co\/#organization","name":"Drone Association Thailand","url":"https:\/\/yousum.gpucore.co\/","logo":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/yousum.gpucore.co\/#\/schema\/logo\/image\/","url":"https:\/\/youzum.net\/wp-content\/uploads\/2024\/11\/tranparent-logo.png","contentUrl":"https:\/\/youzum.net\/wp-content\/uploads\/2024\/11\/tranparent-logo.png","width":300,"height":300,"caption":"Drone Association Thailand"},"image":{"@id":"https:\/\/yousum.gpucore.co\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/DroneAssociationTH\/"]},{"@type":"Person","@id":"https:\/\/yousum.gpucore.co\/#\/schema\/person\/97fa48242daf3908e4d9a5f26f4a059c","name":"admin NU","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/yousum.gpucore.co\/#\/schema\/person\/image\/","url":"https:\/\/youzum.net\/wp-content\/uploads\/avatars\/2\/1746849356-bpfull.png","contentUrl":"https:\/\/youzum.net\/wp-content\/uploads\/avatars\/2\/1746849356-bpfull.png","caption":"admin NU"},"url":"https:\/\/youzum.net\/fr\/members\/adminnu\/"}]}},"rttpg_featured_image_url":null,"rttpg_author":{"display_name":"admin NU","author_link":"https:\/\/youzum.net\/fr\/members\/adminnu\/"},"rttpg_comment":0,"rttpg_category":"<a href=\"https:\/\/youzum.net\/fr\/category\/ai-club\/\" rel=\"category tag\">AI<\/a> <a href=\"https:\/\/youzum.net\/fr\/category\/committee\/\" rel=\"category tag\">Committee<\/a> <a href=\"https:\/\/youzum.net\/fr\/category\/news\/\" rel=\"category tag\">News<\/a> <a href=\"https:\/\/youzum.net\/fr\/category\/uncategorized\/\" rel=\"category tag\">Uncategorized<\/a>","rttpg_excerpt":"Contrastive-LM has released CLM-8B, the first open model in a new class called Contrastive Language Models (CLMs). CLM does not generate text. It scores a set of candidate actions against the current state and returns probabilities. Their main baseline is Jev, the proprietary System One model from TypeSafe AI. Is it deployable? Yes. The Apache-2.0\u2026","_links":{"self":[{"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/posts\/119990","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/comments?post=119990"}],"version-history":[{"count":0,"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/posts\/119990\/revisions"}],"wp:attachment":[{"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/media?parent=119990"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/categories?post=119990"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/youzum.net\/fr\/wp-json\/wp\/v2\/tags?post=119990"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}