Archive
1492 technical news articles
GitHub Repo
Microsoft released ONNX Runtime 1.31.0 on October 9, stabilizing the model package and EPContext data interfaces while improving CPU model loading and quantized MoE execution. The release is available on PyPI, but teams using GPU plugins should verify operator schema compatibility and measure performance in their own environments.
GitHub Repo
An October 8 report from the Ollama community says the Clef reasoning model’s micro-batch size may be set to the full context length, causing memory allocation failures with default settings. The report includes logs from multiple backends and a workaround using a shorter context, but maintainers have yet to confirm or fix the issue.
GitHub Repo
A PyTorch development build report on October 8 found that calculating variance and standard deviation for large-value constant tensors on CPU may incorrectly return infinity. The report includes a minimal reproduction and an algorithm analysis, but the impact on released versions and an upstream fix remain unconfirmed.
最新模型
JetBrains released Mellum2.1 Thinking on October 8. It retains the 12B MoE architecture and focuses this update on reinforcement learning in real software environments. The weights are licensed under Apache 2.0, but agent evaluation results are self-reported, and quantized deployments require separate validation.
GitHub Repo
Ollama released v0.40.1 on October 7, switching Windows model manifests to regular files and adding an on-read repair mechanism for existing symlinks. This gives community reports of previously unusable model names an official release path, though recovery across different Windows environments remains to be verified.
GitHub Repo
An October 8 report from the Open WebUI community says a development build may reject generation requests when a reply message has already been created, preventing automations, sub-agents, and timers from writing answers. The same pattern appears in the documented server-side API flow, but the reporter says the latest stable release works; maintainers have yet to confirm or fix the issue.
最新模型
On October 7, Perplexity announced two open-weight late-interaction models for retrieving text, images, and visual documents. The 0.6B and 9B versions share an embedding space, allowing different-sized encoders to be used for document indexing and online queries, though their actual costs and performance on Chinese remain to be validated.
GitHub Repo
A reproduction on the October 8 Transformers development build found that a uniform frame-sampling function may return more indices than requested, including a position beyond the available frames. The case can be checked without downloading a model, but its impact on stable releases and individual model processors remains unclear.
GitHub Repo
An October 8 reproduction on a PyTorch development build found that Inductor-compiled nll_loss may accept targets with incorrect spatial dimensions and calculate a loss. The same inputs raise an exception in eager and aot_eager, but the impact on released versions and any fix remain unconfirmed.
GitHub Repo
Nous Research released Hermes Agent v0.21.6 on October 8, fixing security issues involving dashboard login and configuration in untrusted repositories. The stable release includes Docker images and a Git tag; the desktop app and Termux package remain on their current versions.
GitHub Repo
Reports filed on October 7 for PyTorch and TorchTitan indicate that copy-on-write tensors materialized during CUDA Graph capture may omit the copy operation, causing incorrect results on replay. The reports include a minimal reproduction and an MoE inference comparison, but the impact on released versions and an upstream fix remain unconfirmed.
GitHub Repo
A development-branch reproduction on October 7 suggests Whisper timestamp decoding may mishandle the two-dimensional tensors returned by `generate`, merging batch results or failing at runtime. The report includes a minimal reproduction, but the impact on released versions and any upstream fix remain unconfirmed.