------------------------------------------------------------------- Sun Aug 11 02:40:06 UTC 2024 - Alessandro de Oliveira Faria - Update to version 0.3.4: * New embedding models - BGE-M3: a large embedding model from BAAI distinguished for its versatility in Multi-Functionality, Multi-Linguality, and Multi-Granularity. - BGE-Large: a large embedding model trained in english. - Paraphrase-Multilingual: A multilingual embedding model trained on parallel data for 50+ languages. * New embedding API with batch support - Ollama now supports a new API endpoint /api/embed for embedding generation: * This API endpoint supports new features: - Batches: generate embeddings for several documents in one request - Normalized embeddings: embeddings are now normalized, improving similarity results - Truncation: a new truncate parameter that will error if set to false - Metrics: responses include load_duration, total_duration and prompt_eval_count metrics ------------------------------------------------------------------- Sat Aug 03 09:41:56 UTC 2024 - eyadlorenzo@gmail.com - Update to version 0.3.3: * The /api/embed endpoint now returns statistics: total_duration, load_duration, and prompt_eval_count * Added usage metrics to the /v1/embeddings OpenAI compatibility API * Fixed issue where /api/generate would respond with an empty string if provided a context * Fixed issue where /api/generate would return an incorrect value for context * /show modefile will now render MESSAGE commands correctly - Update to version 0.3.2: * Fixed issue where ollama pull would not resume download progress * Fixed issue where phi3 would report an error on older versions ------------------------------------------------------------------- Tue Jul 30 07:08:37 UTC 2024 - Adrian Schröter - Update to version 0.3.1: * Added support for min_p sampling option * Lowered number of requests required when downloading models with ollama pull * ollama create will now autodetect required stop parameters when importing certain models * Fixed issue where /save would cause parameters to be saved incorrectly. * OpenAI-compatible API will now return a finish_reason of tool_calls if a tool call occured. ------------------------------------------------------------------- Mon Jul 29 09:59:58 UTC 2024 - Adrian Schröter - fix build on leap 15.6 - exclude builds on 32bit due to build failures ------------------------------------------------------------------- Sun Jul 28 11:32:19 UTC 2024 - Eyad Issa - Update to version 0.3.0: * Ollama now supports tool calling with popular models such as Llama 3.1. This enables a model to answer a given prompt using tool(s) it knows about, making it possible for models to perform more complex tasks or interact with the outside world. * New models: ~ Llama 3.1 ~ Mistral Large 2 ~ Firefunction v2 ~ Llama-3-Groq-Tool-Use * Fixed duplicate error message when running ollama create ------------------------------------------------------------------- Wed Jul 24 14:28:08 UTC 2024 - adrian@suse.de - Update to version 0.2.8: * api embed docs (#5282) * convert: capture `head_dim` for mistral (#5818) * Update llama.cpp submodule commit to `d94c6e0c` (#5805) * server: collect nested tool call objects when parsing (#5824) * Remove no longer supported max vram var * Refine error reporting for subprocess crash * Remove out of space test temporarily (#5825) * llm: consider `head_dim` in llama arch (#5817) * Adjust windows ROCm discovery * add patch for tekken (#5807) * preserve last assistant message (#5802) * Fix generate test flakyness (#5804) * server: validate template (#5734) * OpenAI: Function Based Testing (#5752) * adjust openai chat msg processing (#5729) * fix parsing tool calls * server: check for empty tools array too (#5779) * always provide content even if empty (#5778) * server: only parse tool calls if tools are provided (#5771) * Fix context exhaustion integration test for small gpus * Refine scheduler unit tests for reliability ------------------------------------------------------------------- Thu Jul 18 13:09:10 UTC 2024 - Eyad Issa - Fixed issue with shared libraries ------------------------------------------------------------------- Thu Jul 18 12:27:54 UTC 2024 - Eyad Issa - Added %check section - Use -v when building - Update to version 0.2.6: * New models: MathΣtral is a 7B model designed for math reasoning and scientific discovery by Mistral AI. * Fixed issue where uppercase roles such as USER would no longer work in the chat endpoints * Fixed issue where empty system message would be included in the prompt ------------------------------------------------------------------- Sun Jul 14 17:48:36 UTC 2024 - eyadlorenzo@gmail.com - Update to version 0.2.5: * Fixed issue where a model's SYSTEM message not be applied - Update to version 0.2.4: * Fixed issue where context, load_duration and total_duration fields would not be set in the /api/generate endpoint. * Ollama will no longer error if loading models larger than system memory if disk space is available - Update to version 0.2.3: * Fix issue where system prompt would not be applied - Update to version 0.2.2: * Fixed errors that occurred when using Ollama with Nvidia V100 GPUs * glm4 models will no longer fail to load from out of memory errors * Fixed error that would occur when running deepseek-v2 and deepseek-coder-v2 models * Fixed a series of out of memory issues when using Nvidia GPUs * Fixed a series of errors that would occur when using multiple Radeon GPUs - Update to version 0.2.1: * Fixed issue where setting OLLAMA_NUM_PARALLEL would cause models to be reloaded after each request - Update to version 0.2.0: * Ollama 0.2.0 is now available with concurrency support. This unlocks 2 specific features: ~ Ollama can now serve multiple requests at the same time ~ Ollama now supports loading different models at the same time * New models: GLM-4: A strong multi-lingual general language model with competitive performance to Llama 3. * New models: CodeGeeX4: A versatile model for AI software development scenarios, including code completion. * New models: Gemma 2: Improved output quality and base text generation models now available * Ollama will now show a better error if a model architecture isn't supported * Improved handling of quotes and spaces in Modelfile FROM lines * Ollama will now return an error if the system does not have enough memory to run a model on Linux ------------------------------------------------------------------- Sun Jul 07 19:18:11 UTC 2024 - Eyad Issa - Update to version 0.1.48: * Fixed issue where Gemma 2 would continuously output when reaching context limits * Fixed out of memory and core dump errors when running Gemma 2 * /show info will now show additional model information in ollama run * Fixed issue where ollama show would result in an error on certain vision models - Update to version 0.1.48: * Added support for Google Gemma 2 models (9B and 27B) * Fixed issues with ollama create when importing from Safetensors ------------------------------------------------------------------- Mon Jun 24 10:11:17 UTC 2024 - Eyad Issa - Update to version 0.1.46: * Docs (#5149) * fix: quantization with template * Fix use_mmap parsing for modelfiles * Refine mmap default logic on linux * Bump latest fedora cuda repo to 39 ------------------------------------------------------------------- Sat Jun 22 10:08:00 UTC 2024 - Eyad Issa - Update to version 0.1.45: * New models: DeepSeek-Coder-V2: A 16B & 236B open-source Mixture-of-Experts code language model that achieves performance comparable to GPT4-Turbo in code-specific tasks. * ollama show will now show model information such as context window size * Model loading on Windows with CUDA GPUs is now faster * Setting seed in the /v1/chat/completions OpenAI compatibility endpoint no longer changes temperature * Enhanced GPU discovery and multi-gpu support with concurrency * Introduced a workaround for AMD Vega RX 56 SDMA support on Linux * Fix memory prediction for deepseek-v2 and deepseek-coder-v2 models * api/show endpoint returns extensive model metadata * GPU configuration variables are now reported in ollama serve * Update Linux ROCm to v6.1.1 ------------------------------------------------------------------- Tue Jun 18 12:12:41 UTC 2024 - Eyad Issa - Added documentation files to .spec - Update to version 0.1.44: * Fixed issue where unicode characters such as emojis would not be loaded correctly when running ollama create * Fixed certain cases where Nvidia GPUs would not be detected and reported as compute capability 1.0 devices - Update to version 0.1.43: * New import.md guide for converting and importing models to Ollama * Fixed issue where embedding vectors resulting from /api/embeddings would not be accurate * JSON mode responses will no longer include invalid escape characters * Removing a model will no longer show incorrect File not found errors * Fixed issue where running ollama create would result in an error on Windows with certain file formatting - Update to version 0.1.42: * New models: Qwen 2: a new series of large language models from Alibaba group * Qwen 2: a new series of large language models from Alibaba group * ollama pull is now faster if it detects a model is already downloaded * ollama create will now automatically detect prompt templates for popular model architectures such as Llama, Gemma, Phi and more. * Ollama can now be accessed from local apps built with Electron and Tauri, as well as in developing apps in local html files * Update welcome prompt in Windows to llama3 * Fixed issues where /api/ps and /api/tags would show invalid timestamps in responses - Update to version 0.1.41: * Fixed issue on Windows 10 and 11 with Intel CPUs with integrated GPUs where Ollama would encounter an error ------------------------------------------------------------------- Sat Jun 01 21:12:20 UTC 2024 - Eyad Issa - Update to version 0.1.40: * New model: Codestral: Codestral is Mistral AI’s first-ever code model designed for code generation tasks. * New model: IBM Granite Code: now in 3B and 8B parameter sizes. * New model: Deepseek V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model * Fixed out of memory and incorrect token issues when running Codestral on 16GB Macs * Fixed issue where full-width characters (e.g. Japanese, Chinese, Russian) were deleted at end of the line when using ollama run ------------------------------------------------------------------- Wed May 29 11:38:26 UTC 2024 - Eyad Issa - Update to version 0.1.39: * New model: Cohere Aya 23: A new state-of-the-art, multilingual LLM covering 23 different languages. * New model: Mistral 7B 0.3: A new version of Mistral 7B with initial support for function calling. * New model: Phi-3 Medium: a 14B parameters, lightweight, state-of-the-art open model by Microsoft. * New model: Phi-3 Mini 128K and Phi-3 Medium 128K: versions of the Phi-3 models that support a context window size of 128K * New model: Granite code: A family of open foundation models by IBM for Code Intelligence * It is now possible to import and quantize Llama 3 and its finetunes from Safetensors format to Ollama. * Full changelog at https://github.com/ollama/ollama/releases/tag/v0.1.39 ------------------------------------------------------------------- Wed May 22 18:05:30 UTC 2024 - Eyad Issa - Added 15.6 build ------------------------------------------------------------------- Thu May 16 19:55:51 UTC 2024 - Eyad Issa - Update to version 0.1.38: * New model: Falcon 2: A new 11B parameters causal decoder-only model built by TII and trained over 5T tokens. * New model: Yi 1.5: A new high-performing version of Yi, now licensed as Apache 2.0. Available in 6B, 9B and 34B sizes. * Added ollama ps command * Added /clear command * Fixed issue where switching loaded models on Windows would take several seconds * Running /save will no longer abort the chat session if an incorrect name is provided * The /api/tags API endpoint will now correctly return an empty list [] instead of null if no models are provided ------------------------------------------------------------------- Sun May 12 19:05:53 UTC 2024 - Eyad Issa - Update to version 0.1.37: * Fixed issue where models with uppercase characters in the name would not show with ollama list * Fixed usage string for ollama create * Fix finish_reason being "" instead of null in the Open-AI compatible chat API. ------------------------------------------------------------------- Sun May 12 15:20:28 UTC 2024 - Eyad Issa - Use obs_scm service instead of the deprecated tar_scm - Use zstd for vendor tarball compression ------------------------------------------------------------------- Sun May 12 01:39:26 UTC 2024 - Eyad Issa - Update to version 0.1.36: * Fixed exit status 0xc0000005 error with AMD graphics cards on Windows * Fixed rare out of memory errors when loading a model to run with CPU - Update to version 0.1.35: * New models: Llama 3 ChatQA: A model from NVIDIA based on Llama 3 that excels at conversational question answering (QA) and retrieval-augmented generation (RAG). * Quantization: ollama create can now quantize models when importing them using the --quantize or -q flag * Fixed issue where inference subprocesses wouldn't be cleaned up on shutdown. * Fixed a series out of memory errors when loading models on multi-GPU systems * Ctrl+J characters will now properly add newlines in ollama run * Fixed issues when running ollama show for vision models * OPTIONS requests to the Ollama API will no longer result in errors * Fixed issue where partially downloaded files wouldn't be cleaned up * Added a new done_reason field in responses describing why generation stopped responding * Ollama will now more accurately estimate how much memory is available on multi-GPU systems especially when running different models one after another - Update to version 0.1.34: * New model: Llava Llama 3 * New model: Llava Phi 3 * New model: StarCoder2 15B Instruct * New model: CodeGemma 1.1 * New model: StableLM2 12B * New model: Moondream 2 * Fixed issues with LLaVa models where they would respond incorrectly after the first request * Fixed out of memory errors when running large models such as Llama 3 70B * Fixed various issues with Nvidia GPU discovery on Linux and Windows * Fixed a series of Modelfile errors when running ollama create * Fixed no slots available error that occurred when cancelling a request and then sending follow up requests * Improved AMD GPU detection on Fedora * Improved reliability when using the experimental OLLAMA_NUM_PARALLEL and OLLAMA_MAX_LOADED flags * ollama serve will now shut down quickly, even if a model is loading - Update to version 0.1.33: * New model: Llama 3 * New model: Phi 3 Mini * New model: Moondream * New model: Llama 3 Gradient 1048K * New model: Dolphin Llama 3 * New model: Qwen 110B * Fixed issues where the model would not terminate, causing the API to hang. * Fixed a series of out of memory errors on Apple Silicon Macs * Fixed out of memory errors when running Mixtral architecture models * Aded experimental concurrency features: ~ OLLAMA_NUM_PARALLEL: Handle multiple requests simultaneously for a single model ~ OLLAMA_MAX_LOADED_MODELS: Load multiple models simultaneously ------------------------------------------------------------------- Tue Apr 23 02:26:34 UTC 2024 - rrahl0@disroot.org - Update to version 0.1.32: * scale graph based on gpu count * Support unicode characters in model path (#3681) * darwin: no partial offloading if required memory greater than system * update llama.cpp submodule to `7593639` (#3665) * fix padding in decode * Revert "cmd: provide feedback if OLLAMA_MODELS is set on non-serve command (#3470)" (#3662) * Added Solar example at README.md (#3610) * Update langchainjs.md (#2030) * Added MindsDB information (#3595) * examples: add more Go examples using the API (#3599) * Update modelfile.md * Add llama2 / torch models for `ollama create` (#3607) * Terminate subprocess if receiving `SIGINT` or `SIGTERM` signals while model is loading (#3653) * app: gracefully shut down `ollama serve` on windows (#3641) * types/model: add path helpers (#3619) * update llama.cpp submodule to `4bd0f93` (#3627) * types/model: make ParseName variants less confusing (#3617) * types/model: remove (*Digest).Scan and Digest.Value (#3605) * Fix rocm deps with new subprocess paths * mixtral mem * Revert "types/model: remove (*Digest).Scan and Digest.Value (#3589)" * types/model: remove (*Digest).Scan and Digest.Value (#3589) * types/model: remove DisplayLong (#3587) * types/model: remove MarshalText/UnmarshalText from Digest (#3586) * types/model: init with Name and Digest types (#3541) * server: provide helpful workaround hint when stalling on pull (#3584) * partial offloading * refactor tensor query * api: start adding documentation to package api (#2878) * examples: start adding Go examples using api/ (#2879) * Handle very slow model loads * fix: rope * Revert "build.go: introduce a friendlier way to build Ollama (#3548)" (#3564) * build.go: introduce a friendlier way to build Ollama (#3548) * update llama.cpp submodule to `1b67731` (#3561) * ci: use go-version-file * Correct directory reference in macapp/README (#3555) * cgo quantize * no blob create if already exists * update generate scripts with new `LLAMA_CUDA` variable, set `HIP_PLATFORM` to avoid compiler errors (#3528) * Docs: Remove wrong parameter for Chat Completion (#3515) * no rope parameters * add command-r graph estimate * Fail fast if mingw missing on windows * use an older version of the mac os sdk in release (#3484) * Add test case for context exhaustion * CI missing archive * fix dll compress in windows building * CI subprocess path fix * Fix CI release glitches * update graph size estimate * Fix macOS builds on older SDKs (#3467) * cmd: provide feedback if OLLAMA_MODELS is set on non-serve command (#3470) * feat: add OLLAMA_DEBUG in ollama server help message (#3461) * Revert options as a ref in the server * default head_kv to 1 * fix metal gpu * Bump to b2581 * Refined min memory from testing * Release gpu discovery library after use * Safeguard for noexec * Detect too-old cuda driver * Integration test improvements * Apply 01-cache.diff * Switch back to subprocessing for llama.cpp * Simplify model conversion (#3422) * fix generate output * update memory calcualtions * refactor model parsing * Add chromem-go to community integrations (#3437) * Update README.md (#3436) * Community Integration: CRAG Ollama Chat (#3423) * Update README.md (#3378) * Community Integration: ChatOllama (#3400) * Update 90_bug_report.yml * Add gemma safetensors conversion (#3250) * CI automation for tagging latest images * Bump ROCm to 6.0.2 patch release * CI windows gpu builds * Update troubleshooting link * fix: trim quotes on OLLAMA_ORIGINS - add set_version to automatically switch over to the newer version ------------------------------------------------------------------- Tue Apr 16 10:52:25 UTC 2024 - bwiedemann@suse.com - Update to version 0.1.31: * Backport MacOS SDK fix from main * Apply 01-cache.diff * fix: workflows * stub stub * mangle arch * only generate on changes to llm subdirectory * only generate cuda/rocm when changes to llm detected * Detect arrow keys on windows (#3363) * add license in file header for vendored llama.cpp code (#3351) * remove need for `$VSINSTALLDIR` since build will fail if `ninja` cannot be found (#3350) * change `github.com/jmorganca/ollama` to `github.com/ollama/ollama` (#3347) * malformed markdown link (#3358) * Switch runner for final release job * Use Rocky Linux Vault to get GCC 10.2 installed * Revert "Switch arm cuda base image to centos 7" * Switch arm cuda base image to centos 7 * Bump llama.cpp to b2527 * Fix ROCm link in `development.md` * adds ooo to community integrations (#1623) * Add cliobot to ollama supported list (#1873) * Add Dify.AI to community integrations (#1944) * enh: add ollero.nvim to community applications (#1905) * Add typechat-cli to Terminal apps (#2428) * add new Web & Desktop link in readme for alpaca webui (#2881) * Add LibreChat to Web & Desktop Apps (#2918) * Add Community Integration: OllamaGUI (#2927) * Add Community Integration: OpenAOE (#2946) * Add Saddle (#3178) * tlm added to README.md terminal section. (#3274) * Update README.md (#3288) * Update README.md (#3338) * Integration tests conditionally pull * add support for libcudart.so for CUDA devices (adds Jetson support) * llm: prevent race appending to slice (#3320) * Bump llama.cpp to b2510 * Add Testcontainers into Libraries section (#3291) * Revamp go based integration tests * rename `.gitattributes` * Bump llama.cpp to b2474 * Add docs for GPU selection and nvidia uvm workaround * doc: faq gpu compatibility (#3142) * Update faq.md * Better tmpdir cleanup * Update faq.md * update `faq.md` * dyn global * llama: remove server static assets (#3174) * add `llm/ext_server` directory to `linguist-vendored` (#3173) * Add Radeon gfx940-942 GPU support * Wire up more complete CI for releases * llm,readline: use errors.Is instead of simple == check (#3161) * server: replace blob prefix separator from ':' to '-' (#3146) * Add ROCm support to linux install script (#2966) * .github: fix model and feature request yml (#3155) * .github: add issue templates (#3143) * fix: clip memory leak * Update README.md * add `OLLAMA_KEEP_ALIVE` to environment variable docs for `ollama serve` (#3127) * Default Keep Alive environment variable (#3094) * Use stdin for term discovery on windows * Update ollama.iss * restore locale patch (#3091) * token repeat limit for prediction requests (#3080) * Fix iGPU detection for linux * add more docs on for the modelfile message command (#3087) * warn when json format is expected but not mentioned in prompt (#3081) * Adapt our build for imported server.cpp * Import server.cpp as of b2356 * refactor readseeker * Add docs explaining GPU selection env vars * chore: fix typo (#3073) * fix gpu_info_cuda.c compile warning (#3077) * use `-trimpath` when building releases (#3069) * relay load model errors to the client (#3065) * Update troubleshooting.md * update llama.cpp submodule to `ceca1ae` (#3064) * convert: fix shape * Avoid rocm runner and dependency clash * fix `03-locale.diff` * Harden for deps file being empty (or short) * Add ollama executable peer dir for rocm * patch: use default locale in wpm tokenizer (#3034) * only copy deps for `amd64` in `build_linux.sh` * Rename ROCm deps file to avoid confusion (#3025) * add `macapp` to `.dockerignore` * add `bundle_metal` and `cleanup_metal` funtions to `gen_darwin.sh` * tidy cleanup logs * update llama.cpp submodule to `77d1ac7` (#3030) * disable gpu for certain model architectures and fix divide-by-zero on memory estimation * Doc how to set up ROCm builds on windows * Finish unwinding idempotent payload logic * update llama.cpp submodule to `c2101a2` (#3020) * separate out `isLocalIP` * simplify host checks * add additional allowed hosts * Update docs `README.md` and table of contents * add allowed host middleware and remove `workDir` middleware (#3018) * decode ggla * convert: fix default shape * fix: allow importing a model from name reference (#3005) * update llama.cpp submodule to `6cdabe6` (#2999) * Update api.md * Revert "adjust download and upload concurrency based on available bandwidth" (#2995) * cmd: tighten up env var usage sections (#2962) * default terminal width, height * Refined ROCm troubleshooting docs * Revamp ROCm support * update go to 1.22 in other places (#2975) * docs: Add LLM-X to Web Integration section (#2759) * fix some typos (#2973) * Convert Safetensors to an Ollama model (#2824) * Allow setting max vram for workarounds * cmd: document environment variables for serve command * Add Odin Runes, a Feature-Rich Java UI for Ollama, to README (#2440) * Update api.md * Add NotesOllama to Community Integrations (#2909) * Added community link for Ollama Copilot (#2582) * use LimitGroup for uploads * adjust group limit based on download speed * add new LimitGroup for dynamic concurrency * refactor download run ------------------------------------------------------------------- Wed Mar 06 23:51:28 UTC 2024 - computersemiexpert@outlook.com - Update to version 0.1.28: * Fix embeddings load model behavior (#2848) * Add Community Integration: NextChat (#2780) * prepend image tags (#2789) * fix: print usedMemory size right (#2827) * bump submodule to `87c91c07663b707e831c59ec373b5e665ff9d64a` (#2828) * Add ollama user to video group * Add env var so podman will map cuda GPUs ------------------------------------------------------------------- Tue Feb 27 08:33:15 UTC 2024 - Jan Engelhardt - Edit description, answer _what_ the package is and use nominal phrase. (https://en.opensuse.org/openSUSE:Package_description_guidelines) ------------------------------------------------------------------- Fri Feb 23 21:13:53 UTC 2024 - Loren Burkholder - Added the Ollama package - Included a systemd service