Skip to main content
← All Tags

Language Model

49 articles in this category (Page 2 of 3)

AI NewsLanguage ModelApplications

A Coding Implementation to Automating LLM Quality Assurance with DeepEval, Custom Retrievers, and LLM-as-a-Judge Metrics

This article details a coding implementation for automated LLM quality assurance, achieving rigorous testing through DeepEval, custom retrievers, and LLM-as-a-judge metrics.

Read more
AI NewsAI AgentsLanguage Model

Zhipu AI Releases GLM-4.7-Flash: A 30B-A3B MoE Model for Efficient Local Coding and Agents

Zhipu AI released GLM-4.7-Flash, a 31B parameter Mixture of Experts model achieving leading performance among 30B models on coding and reasoning benchmarks.

Read more
AI NewsLanguage ModelOpen Source

Nous Research Releases NousCoder-14B: A Competitive Olympiad Programming Model

Nous Research’s NousCoder-14B achieves a 67.87% Pass@1 accuracy on the LiveCodeBench v6 benchmark, surpassing the Qwen3-14B baseline by 7.08 percentage points.

Read more
AI NewsLanguage ModelArtificial Intelligence

NVIDIA Releases PersonaPlex-7B-v1: A Real-Time Speech-to-Speech Model

NVIDIA’s PersonaPlex-7B-v1 achieves a 0.908 Takeover Rate on FullDuplexBench, demonstrating significant progress in natural, full-duplex conversational AI.

Read more
AI NewsLanguage ModelMachine Learning

TII Abu-Dhabi Released Falcon H1R-7B: A New Reasoning Model Outperforming Others in Math and Coding

Technology Innovation Institute (TII) released Falcon-H1R-7B, a 7B parameter model achieving performance comparable to 14B-47B models in math, code, and reasoning benchmarks.

Read more
AI NewsLanguage ModelOpen Source

LLM-Pruning Collection: A JAX Framework for LLM Compression

Researchers released LLM-Pruning Collection, a JAX-based repository consolidating major pruning algorithms for large language models, aiming to standardize comparison and reproducibility.

Read more
AI NewsLanguage ModelAI Paper Summary

Tencent Releases HY-MT1.5 Translation Models: 1.8B & 7B Parameters for Cloud & Edge

Tencent’s HY-MT1.5 translation models achieve industry-leading performance, with the 1.8B version running on 1GB of memory with 0.18-second latency.

Read more
AI NewsAI AgentsLanguage Model

Recursive Language Models (RLMs): From MIT’s Blueprint to Prime Intellect’s RLMEnv for Long Horizon LLM Agents

Recursive Language Models (RLMs) achieve up to 62% accuracy on CodeQA, significantly improving upon standard LLM performance and reducing query costs.

Read more
AI NewsLanguage ModelArtificial Intelligence

FunctionGemma: Google AI’s 270M Parameter Function Calling Specialist for Edge Workloads

Google released FunctionGemma, a compact 270M parameter model achieving 85% accuracy on the Mobile Actions benchmark after fine-tuning.

Read more
AI NewsArtificial IntelligenceLanguage Model

Google Health AI Releases MedASR: A Conformer-Based Medical Speech-to-Text Model

Google released MedASR, a 105M parameter medical speech-to-text model, achieving up to 4.6% word error rate in radiology dictation with a language model.

Read more
AI NewsLanguage ModelMachine Learning

Google Introduces T5Gemma 2: Encoder Decoder Models with Multimodal Inputs via SigLIP and 128K Context

Google released T5Gemma 2, a family of open-source encoder-decoder models inheriting Gemma 3’s multimodality and 128K context length.

Read more
AI NewsLanguage ModelComputer Vision

Zhipu AI Releases GLM-4.6V: A 128K Context Vision Language Model with Native Tool Calling

Zhipu AI launched GLM-4.6V, a 106B parameter multimodal model with a 128K token context window, enabling native multimodal function calling for improved agent capabilities.

Read more
AI NewsLanguage ModelMachine Learning

Apple Researchers Release CLaRa: A Continuous Latent Reasoning Framework for Compression-Native RAG with 16x–128x Semantic Document Compression

Apple's CLaRa achieves 16x–128x semantic document compression, boosting RAG efficiency without sacrificing accuracy.

Read more
AI NewsLanguage ModelLarge Language Model

NVIDIA and Mistral AI Bring 10x Faster Inference for the Mistral 3 Family on GB200 NVL72 GPU Systems

NVIDIA and Mistral AI achieve 10x faster inference for Mistral 3 models on GB200 NVL72 GPUs, reaching 5M tokens per second per MW.

Read more
AI NewsAudio Language ModelLanguage Model

StepFun AI Releases Step-Audio-R1: A New Audio LLM that Finally Benefits from Test Time Compute Scaling

StepFun AI’s Step-Audio-R1 achieves 83.6% accuracy on audio benchmarks by addressing training limitations, not audio modality flaws.

Read more
AI NewsLanguage ModelTutorials

An Implementation of Fully Traced and Evaluated Local LLM Pipeline Using Opik

This tutorial details building a fully traced LLM pipeline with Opik, achieving transparent, measurable, and reproducible AI workflows with a 95% accuracy score.

Read more
AI NewsLanguage ModelOpen Source

Allen Institute for AI (AI2) Introduces Olmo 3: Open Source 7B/32B LLMs with 65K Context Window

Allen Institute for AI (AI2) launches Olmo 3, open-source 7B/32B LLMs with 65,536 token context window and Dolma 3 data stack.

Read more
AI NewsLanguage ModelMachine Learning

NVIDIA AI Introduces TiDAR: A Hybrid Diffusion Autoregressive Architecture For High Throughput LLM Inference

NVIDIA's TiDAR achieves 5.91x speedup on 8B models while maintaining autoregressive quality.

Read more
AI NewsVoice AILanguage Model

Meta AI Releases Omnilingual ASR: A Suite of Open-Source Multilingual Speech Recognition Models for 1600+ Languages

Meta AI launches Omnilingual ASR, an open-source speech recognition system supporting 1600+ languages with <10% character error rate.

Read more
AI NewsAgentic AILanguage Model

Gelato-30B-A3B: A State-of-the-Art Grounding Model for GUI Computer-Use Tasks, Surpassing Computer Grounding Models like GTA1-32B

Gelato-30B-A3B achieves 63.88% accuracy on ScreenSpot Pro, outperforming GTA1-32B and larger VLMs in GUI grounding tasks.

Read more
AI NewsArtificial IntelligenceLanguage Model

Google AI Introduces Consistency Training for Safer Language Models Under Sycophantic and Jailbreak Style Prompts

Google AI introduces Consistency Training (Bias Augmented Consistency Training and Activation Consistency Training) to enhance language models' safety against sycophantic and jailbreak prompts while preserving their capabilities.

Read more
AI NewsApplicationsArtificial Intelligence

Comparing the Top 7 Large Language Models LLMs/Systems for Coding in 2025

Compare the top 7 large language models and systems for coding in 2025. Discover which ones excel for software engineering tasks.

Read more
AI NewsAgentic AIArtificial Intelligence

Anthropic's Research Demonstrates Claude's Introspective Awareness Through Concept Injection in Controlled Layers

Anthropic's study reveals that Claude models can detect injected concepts via internal activations, offering causal evidence of introspection. The research highlights controlled success rates and implications for LLM transparency.

Read more
AI NewsArtificial IntelligenceLanguage Model

Google AI Unveils Supervised Reinforcement Learning (SRL): A Step-Wise Framework for Enhancing Small Language Models

Google AI introduces Supervised Reinforcement Learning (SRL), a novel training framework that improves small language models' reasoning capabilities by leveraging expert trajectories and step-wise reward mechanisms.

Read more