1676 articles tagged with Benchmark
Guardrail Technologies Releases New Benchmark Report Tracking How the S&P 500 Discloses AI Cybersecurity Risk New research finds 97 percent of S&P 500
Information retrieval (IR) increasingly targets open-ended queries that admit diverse perspectives. Existing IR benchmarks, however, focus primarily on closed-ended queries, while ...
Voice agents fail on latency before they fail on intelligence. A new benchmark tests every layer of the voice stack - speech-to-text, language models, and text-to-speech - revealin...
PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX#PTX #LLM #Packagehttps://hgpu.org/?p=31135
RealisticTritonBench: A Benchmark for Triton-Kernel Generation in Real-World AI Frameworks#Triton #CUDA #LLM #Packagehttps://hgpu.org/?p=31131
Check characters: "🤗 Qwen/Qwen3.8-Flash-Next climbing on Hugging Face: 52.3k downloads/30dhttps://olud.ai/#leaderboard#OpenSource #AI #HuggingFace
Muse Spark 1.2 Contributor (Meta) just hit the leaderboard with a 1M context window at $0.1/$0.2 per million tokens. That pricing for that context length is worth a look.https://ol...
An open-weight AI model just beat GPT-5.6 Luna on a major benchmark. 🤯GLM-5.3-Flash: 57GPT-5.6 Luna: 52GLM-5.3-Flash also offers:🔓 Open weights📚 1M-token context💰 Lower costOne ben...
【QIMMA قِمّة ⛰: 品質第一のアラビア語LLMリーダーボード】https://huggingface.co/blog/tiiuae/qimma-arabic-leaderboard※AI生成の自動投稿（見出し＋リンク）#AI #生成AI #LLM #AIGenerated
【Open ASR リーダーボードに Benchmaxxer Repellant を追加】https://huggingface.co/blog/open-asr-leaderboard-private-data※AI生成の自動投稿（見出し＋リンク）#AI #生成AI #LLM #AIGenerated
Ox Alpha topped OpenRouter's leaderboard under an anonymous name last week. Bloomberg and TechCrunch traced it to Z.ai, maker of GLM.A sudden #1 finish under a fake name is not evi...
📦 Fresh open weights on the leaderboard: Hy-MT2-1.8B (Tencent)Context: 8.2k tokens · $0.04 in / $0.18 out per 1M · open weightsAll the latest models, tracked hourly:https://olud.ai...
RT @0xSero: Deepseek-v4-Flash-exl3-reap mehr auf Arint.info #AI #Benchmark #Deepseek #LLM #MachineLearning #Tech #arint_info https://x.com/0xSero/status/2093754108708598240
📊 Molmo 7B-D — the actual numbers GPQA: 24% MMLU-Pro: 37.1% Humanity's Last Exam: 5.1% Long Context Reasoning: 0%Measured independently, not self-reported →https://olud.ai/leaderbo...
Bench'd (benchd.ai) calls itself the neutral benchmark authority for AI memory, and sells vendors a...
A 4.6-point gap in the leaderboard, but 200x cheaper on output: GLM 5.3 Flash makes you ask what you're really paying for with Claude Fable 5.https://olud.ai/leaderboard.html#OpenS...
📊 Mistral Medium 3 — the actual numbers GPQA: 57.8% MMLU-Pro: 76% Humanity's Last Exam: 4.1% Long Context Reasoning: 31.7%💰 15.6 intelligence points per dollarMeasured independentl...
【Open ASRリーダーボードに初のグローバルサウス言語が追加される】https://huggingface.co/blog/open-asr-leaderboard-global-south※AI生成の自動投稿（見出し＋リンク）#AI #生成AI #LLM #AIGenerated
Benchmark-driven configuration optimization for vLLM inference servers
RT @googlegemma: 3.431 Token/s auf Gemma 4 31B! mehr auf Arint.info #AI #Benchmark #Gemma #Groq #LLM #NVIDIA #arint_info https://x.com/googlegemma/status/2092666213914771529
A prompt regression is the only production bug that never throws an exception and never writes to...
Your evaluation set has an expiration date. Last month it helped you pick a model. This month it is...
you can't use nVidia performance to benchmark OpenAI and Anthropic. For them the question is: does the price that they can sell inference for exceed the cost of that inference? If ...