News
News
OpenAI o4-proto-6 enables 96-hour autonomous ML model fine-tuning and deployment loop
[AI-NEWS]...
2026-02-28
Mistral Pixtral Large 2.5 reaches SOTA open multimodal reasoning on MMMU-Pro and MathVista-Hard
Content : Pixtral Large 2.5 (open-weight) sets new open-model records on MMMU-Pro (visual reasoning) and MathVista-Hard (math + vision),
2026-02-25
Anthropic Claude 5 Sonnet demonstrates 14-day autonomous scientific literature review + hypothesis generation
Claude 5 Sonnet completes 14-day autonomous loop:
2026-02-25
xAI Grok-5.1 integrates real-time multi-agent debate for improved truthfulness and calibration
Grok-5.1 introduces native multi-agent debate layer
2026-02-25
Google DeepMind AlphaFold 4 predicts all known protein complexes with near-experimental accuracy
AlphaFold 4 reaches near-experimental RMSD (<1.0 Å) on nearly all solved protein complexes in PDB
2026-02-25
OpenAI o4-proto-4 achieves 82% autonomous resolution on full SWE-Bench Verified + Multi-File
[AI-NEWS]...
2026-02-25
Mistral Mathstral 2.0 sets new open-source SOTA on MATH and GSM8K-Hard
Mistral Mathstral 2.0 achieves 94.1% on MATH and 98.7% on GSM8K-Hard, new open-source SOTA
2026-02-22
Anthropic Claude 4.4 Opus demonstrates 7-day autonomous research agent on novel scientific hypothesis
Claude 4.4 Opus demonstrates 7-day autonomous research agent producing publication-quality chemistry papers
2026-02-22
xAI Grok-5 preview adds native causal video generation with physics consistency
xAI Grok-5 preview adds causal video generation with physics consistency for 10-second 720p clips
2026-02-22
OpenAI o4-proto-3 achieves closed-loop self-correction on 100+ cycle software debugging tasks
OpenAI o4-proto-3 achieves closed-loop self-correction with 74.2% resolution on 100+ cycle debugging tasks
2026-02-22
Mistral Devstral: Specialized 24B Model Optimized for Developer Tool-Chain Integration
Mistral released Devstral (24B), a developer-focused model fine-tuned for seamless integration with IDEs, terminals, git
2026-02-20
Meta Llama 4 Maverick: First Open Model to Reach Human-Parity on SWE-Bench Verified
Llama 4 Maverick (open-weight 405B variant) achieves 68.7% resolution rate on SWE-Bench Verified
2026-02-20