Ollama v0.1.9 — Release Notes & Engineering Impact
Ollama v0.1.9. Source highlight: Yi: a high-performing, bilingual model supporting both English and Chinese.
AI Models & LLM Frameworks archive — Page 20
1172 tracked entries. Older history is imported progressively; see source coverage.
Dify: v0.3.29
Dify 0.3.29. Source highlight: App Moderation Support [DOCS WIP]
Ollama v0.1.8 — Release Notes & Engineering Impact
Ollama v0.1.8. Source highlight: CodeBooga: A high-performing code instruct model created by merging two existing code models.
Hugging Face Transformers: Safetensors serialization by default, DistilWhisper, Fuyu, Kosmos-2, SeamlessM4T, Owl-v2
Hugging Face Transformers v4.35.0. Source highlight: Distil-Whisper is a distilled version of Whisper that is 6 times faster, 49% smaller, and performs within 1% word error rate (WER) on out-of-distribution data. It was proposed in the paper Robust Knowledge Distillation via Large-Scale Pseudo Labelling.
LocalAI v1.40.0 — Release Notes & Engineering Impact
LocalAI v1.40.0. Source highlight: This release is a preparation before v2 - the efforts now will be to refactor, polish and add new backends. Follow up on: https://github.com/mudler/LocalAI/issues/1126
Ollama v0.1.7 — Release Notes & Engineering Impact
Ollama v0.1.7. Source highlight: Fixed an issue when running `ollama run` where certain key combinations such as Ctrl+Space would lead to an unresponsive prompt
Ollama v0.1.6 — Release Notes & Engineering Impact
Ollama v0.1.6. Source highlight: Dolphin 2.1 Mistral: an instruct-tuned model based on Mistral and trained on a dataset filtered to remove alignment and bias.
Ollama v0.1.5 — Release Notes & Engineering Impact
Ollama v0.1.5. Source highlight: Fix an issue where an error would occur when running `falcon` or `starcoder` models
Ollama v0.1.4 — Release Notes & Engineering Impact
Ollama v0.1.4. Source highlight: OpenHermes 2 Mistral: a new fine-tuned model based on Mistral, trained on open datasets totalling over 900,000 instructions. This model has strong multi-turn chat skills, surpassing previous Hermes 13B models and even matching 70B models on some benchmarks.
Hugging Face Transformers: Patch release: v4.34.1
Hugging Face Transformers v4.34.1. Source highlight: A patch release was made for the following three commits:
vLLM v0.2.1.post1 — Release Notes & Engineering Impact
vLLM v0.2.1.post1 is published, but the available notes are too limited to assess upgrade impact. Review the official source before upgrading.
vLLM v0.2.1 — Release Notes & Engineering Impact
vLLM v0.2.1. Source highlight: PagedAttention V2 kernel: Up to 20% end-to-end latency reduction
Dify: v0.3.28
Dify 0.3.28. Source highlight: fix: new app with template display
Ollama v0.1.3 — Release Notes & Engineering Impact
Ollama v0.1.3. Source highlight: Improved various API error messages to be easier to read
Ollama v0.1.2 — Release Notes & Engineering Impact
Ollama v0.1.2. Source highlight: Zephyr A fine-tuned 7B version of mistral that was trained on a mix of publicly available, synthetic datasets and performs as well as Llama 2 70B in many benchmarks
Dify: v0.3.27
Dify 0.3.27. Source highlight: feat: using random sampling to check if it violates the review mechan…
Dify: v0.3.26
Dify 0.3.26. Source highlight: document segmentApi Add get&update&delete operate
Dify: v0.3.25
Dify 0.3.25. Source highlight: fix: request timeout when openai completion
Hugging Face Transformers: v4.34: Mistral, Persimmon, Prompt templating, Flash Attention 2, Tokenizer refactor
Hugging Face Transformers v4.34.0. Source highlight: Mistral-7B-v0.1 is a decoder-based LM with the following architectural choices:
Ollama v0.1.1 — Release Notes & Engineering Impact
Ollama v0.1.1. Source highlight: Cancellable responses: `Ctrl+C` will now cancel responses when running `ollama run`
vLLM v0.2.0 — Release Notes & Engineering Impact
vLLM v0.2.0. Source highlight: Up to 60% performance improvement by optimizing de-tokenization and sampler
Dify: v0.3.24
Dify 0.3.24. Source highlight: add help wanted issue template
Hugging Face Transformers: Patch release: v4.33.3
Hugging Face Transformers v4.33.3. Source highlight: A patch release was made for the following three commits:
LocalAI v1.30.0 — Release Notes & Engineering Impact
LocalAI v1.30.0. Source highlight: This is an exciting LocalAI release! Besides bug-fixes and enhancements this release brings the new backend to a whole new level by extending support to `vllm` and `vall-e-x` for audio generation!
Ollama v0.1.0 — Release Notes & Engineering Impact
Ollama v0.1.0. Source highlight: Ollama for Linux is now available, with GPU acceleration enabled out-of-the-box for Nvidia GPUs.
Ollama v0.0.21 — Release Notes & Engineering Impact
Ollama v0.0.21. Source highlight: Fixed an issue where empty responses would be returned if `template` was provided in the api, but not `prompt`
Ollama v0.0.20 — Release Notes & Engineering Impact
Ollama v0.0.20. Source highlight: `ollama run` has a new & improved experience:
Dify: Version 0.3.23
Dify 0.3.23. Source highlight: feat: spark free quota verify
Hugging Face Transformers: Patch release: v4.33.2
Hugging Face Transformers v4.33.2. Source highlight: A patch release was done for these two commits:
Ollama v0.0.19 — Release Notes & Engineering Impact
Ollama v0.0.19. Source highlight: Updated Docker image for Ollama `docker pull ollama/ollama`
vLLM v0.1.7 — Release Notes & Engineering Impact
vLLM v0.1.7. Source highlight: A minor release to fix the bugs in ALiBi, Falcon-40B, and Code Llama.
Dify: Version 0.3.22
Dify 0.3.22. Source highlight: Notice: if you're upgrading from before version 0.3.21, make sure you've read the **BREAKING CHANGE** in version 0.3.21: https://github.com/langgenius/dify/releases/tag/0.3.21
Dify: Version 0.3.21
Dify 0.3.21. Source highlight: ⚠️ [BREAKING CHANGE] ⚠️
Dify: Version 0.3.20
Dify 0.3.20. Source highlight: feat: optimize generation of conversation title
vLLM v0.1.6 — Release Notes & Engineering Impact
vLLM v0.1.6. Source highlight: **Note:** This is an emergency release to revert a breaking API change that can make many existing codes using AsyncLLMServer not work.
vLLM v0.1.5 — Release Notes & Engineering Impact
vLLM v0.1.5. Source highlight: Align beam search with `hf_model.generate`.
Hugging Face Transformers: Falcon, Code Llama, ViTDet, DINO v2, VITS
Hugging Face Transformers v4.33.1. Source highlight: Falcon is a class of causal decoder-only models built by TII. The largest Falcon checkpoints have been trained on >=1T tokens of text, with a particular emphasis on the RefinedWeb corpus. They are made available under the Apache 2.0 license.
Ollama v0.0.18 — Release Notes & Engineering Impact
Ollama v0.0.18. Source highlight: New `ollama show` command for viewing details about a model:
Dify: Version 0.3.19
Dify 0.3.19. Source highlight: document limit
Ollama v0.0.17 — Release Notes & Engineering Impact
Ollama v0.0.17. Source highlight: Multiple models can be removed together: `ollama rm mario:latest orca-mini:3b`
Hugging Face Transformers: Patch release: v4.32.1
Hugging Face Transformers v4.32.1. Source highlight: Patch release including several patches from v4.31.0, listed below:
LocalAI v1.25.0 — Release Notes & Engineering Impact
LocalAI v1.25.0. Source highlight: feat: bump llama.cpp, add gguf support
Ollama v0.0.16 — Release Notes & Engineering Impact
Ollama v0.0.16. Source highlight: Ollama version can be checked by running `ollama -v` or `ollama --version`
vLLM v0.1.4
vLLM v0.1.4. Source highlight: From now on, vLLM is published with pre-built CUDA binaries. Users don't have to compile the vLLM's CUDA kernels on their machine.
Dify: Version 0.3.18
Dify 0.3.18. Source highlight: feat: remove unnecessary prompt of baichuan
Dify: Version 0.3.17
Dify 0.3.17. Source highlight: feat: adjust hf max tokens
Dify: Version 0.3.16
Dify 0.3.16. Source highlight: normalize embedding
Dify: Version 0.3.15
Dify 0.3.15. Source highlight: feat: add spark v2 support
Hugging Face Transformers: IDEFICS, GPTQ Quantization
Hugging Face Transformers v4.32.0. Source highlight: The IDEFICS model was proposed in OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents by Hugo Laurençon, Lucile Saulnier, Léo Tronchon, Stas Bekman, Amanpreet Singh, Anton Lozhkov, Thomas Wang, Siddharth Karamcheti, Alexander M. Rush, Douwe Kiela, Matthieu Cord, Victor Sa
Ollama v0.0.15 — Release Notes & Engineering Impact
Ollama v0.0.15. Source highlight: 📍 Ollama model list is now available