[Monitor] Frontier update — week of 2026-08-17

[github-actions]

🔭 Frontier model & benchmark monitor — 2026-08-17

Auto-generated by monitor_models.py · lookback: 7 days

🤖 New open-weight model releases (Hugging Face Hub)

• deepseek-ai/DeepSeek-V4-Pro-0813 — deepseek-ai · ⬇️ 21,873 · ❤️ 539
  • Tags: transformers, safetensors, deepseek_v4, text-generation, conversational
• Qwen/Qwen3.8-27B — Qwen · ⬇️ 267,725 · ❤️ 10316
  • Tags: transformers, safetensors, qwen3_5, image-text-to-text, conversational
• Qwen/Qwen3.8-27B-FP8 — Qwen · ⬇️ 352,971 · ❤️ 488
  • Tags: transformers, safetensors, qwen3_5, image-text-to-text, conversational
• Qwen/Qwen3.8-2.4T-A95B — Qwen · ⬇️ 7,932 · ❤️ 1012
  • Tags: transformers, safetensors, qwen3_5_moe_text, text-generation, conversational
• Qwen/Qwen3.8-2.4T-A95B-FP8 — Qwen · ⬇️ 11,311 · ❤️ 210
  • Tags: transformers, safetensors, qwen3_5_moe_text, text-generation, conversational
• microsoft/Fara-7B — microsoft · ⬇️ 1,816 · ❤️ 619
  • Tags: transformers, safetensors, qwen2_5_vl, image-text-to-text, multimodal
• microsoft/Mage-VL — microsoft · ⬇️ 485,557 · ❤️ 347
  • Tags: transformers, safetensors, mage_vl, image-text-to-text, multimodal
• nvidia/NVIDIA-Nemotron-Labs-Teacher-Instruction-Following — nvidia · ⬇️ 806 · ❤️ 0
  • Tags: transformers, safetensors, nemotron_h, text-generation, nvidia
• nvidia/NVIDIA-Nemotron-Labs-Teacher-Competition-Coding — nvidia · ⬇️ 1,231 · ❤️ 4
  • Tags: transformers, safetensors, nemotron_h, text-generation, nvidia
• nvidia/NVIDIA-Nemotron-Labs-Teacher-Chat — nvidia · ⬇️ 1,128 · ❤️ 0
  • Tags: transformers, safetensors, nemotron_h, text-generation, nvidia
• nvidia/Nemotron-Labs-Audex-30B-A3B — nvidia · ⬇️ 1,217 · ❤️ 174
  • Tags: transformers, safetensors, nemotron_labs_audex, nvidia, nemotron-labs-audex
• internlm/Intern-S2-Mobius — internlm · ⬇️ 1,128 · ❤️ 51
  • Tags: transformers, safetensors, interns2_mobius, image-text-to-text, conversational

📄 New model / evaluation papers (arXiv cs.CL + cs.AI)

No new model release papers found this week.

🆕 New benchmarks detected (not in benchmark_registry.json)

No new benchmarks detected beyond our current registry.

🏆 Papers With Code SOTA updates

No new SOTA entries this week.

✅ Suggested actions

• Review new HF models above — add any to results/benchmark.csv as new columns
• Check new arXiv papers for benchmark scores to add to frontier baselines
• Evaluate new benchmarks against your H100 pipeline compatibility
• Update docs/benchmark_registry.json if new benchmarks are confirmed valuable

---

Triggered by .github/workflows/monitor.yml · View workflow