@ -1,14 +1,76 @@
-------------------------------------------------------------------
Thu Nov 07 12:06:09 UTC 2024 - adrian@suse.de
- Update to version 0.4.0:
* Update README.md (#7516)
* One corrupt manifest should not wedge model operations (#7515)
* prompt: Use a single token when estimating mllama context size
* readme: add Hexabot to the list of community integrations
* Quiet down debug log of image payload (#7454)
-------------------------------------------------------------------
Wed Nov 06 12:31:53 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- Update to version 0.4.0-rc8:
* CI: Switch to v13 macos runner (#7498)
* CI: matrix strategy fix (#7496)
* Sign windows arm64 official binaries (#7493)
* readme: add TextCraft to community integrations (#7377)
* nvidia libs have inconsistent ordering (#7473)
* CI: omit unused tools for faster release builds (#7432)
* llama: Improve error handling
* runner.go: Only allocate 1 element embedding batches for mllama
* refactor kv estimation
* mllama cross attention
* Add basic mllama integration tests (#7455)
* runner.go: Don't set cross attention before sending embeddings
* Give unicode test more time to run (#7437)
-------------------------------------------------------------------
Fri Nov 01 02:18:50 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- Remove enable-lto.patch
- Update to version 0.4.0-rc6:
* Refine default thread selection for NUMA systems (#7322)
* runner.go: Better abstract vision model integration
* Soften windows clang requirement (#7428)
* Remove submodule and shift to Go server - 0.4.0 (#7157)
* Move windows app out of preview (#7347)
* windows: Support alt install paths, fit and finish (#6967)
* add more tests for getting the optimal tiled canvas (#7411)
* Switch windows to clang (#7407)
* tests: Add test for Unicode processing
* runner.go: Better handle return NULL values from llama.cpp
* add mllama image processing to the generate handler (#7384)
* Bump to latest Go 1.22 patch (#7379)
* Fix deepseek deseret regex (#7369)
* Better support for AMD multi-GPU on linux (#7212)
* Fix unicode output on windows with redirect to file (#7358)
* Fix incremental build file deps (#7361)
* Improve dependency gathering logic (#7345)
* fix #7247 - invalid image input (#7249)
* integration: harden embedding test (#7306)
* default to "FROM ." if a Modelfile isn't present (#7250)
* Fix rocm windows build and clean up dependency gathering (#7305)
* runner.go: Merge partial unicode characters before sending
* readme: add Ollama for Swift to the community integrations (#7295)
* server: allow vscode-webview origin (#7273)
* image processing for llama3.2 (#6963)
* llama: Decouple patching script from submodule (#7139)
* llama: add compiler tags for cpu features (#7137)
-------------------------------------------------------------------
Wed Oct 30 01:47:37 UTC 2024 - Alessandro de Oliveira Faria <cabelo@opensuse.org>
- Update to version 0.3.14:
* New Models
+ Granite 3 MoE: The IBM Granite 1B and 3B models are the
first mixture of experts (MoE) Granite models from IBM
+ Granite 3 MoE: The IBM Granite 1B and 3B models are the
first mixture of experts (MoE) Granite models from IBM
designed for low latency usage.
+ Granite 3 Dense: The IBM Granite 2B and 8B models are
designed to support tool-based use cases and support for
retrieval augmented generation (RAG), streamlining code
+ Granite 3 Dense: The IBM Granite 2B and 8B models are
designed to support tool-based use cases and support for
retrieval augmented generation (RAG), streamlining code
generation, translation and bug fixing.
-------------------------------------------------------------------
@ -19,21 +81,21 @@ Sat Oct 12 20:55:18 UTC 2024 - eyadlorenzo@gmail.com
~ Llama Guard 3: a series of models by Meta, fine-tuned for
content safety classification of LLM inputs and responses.
~ ShieldGemma: ShieldGemma is set of instruction tuned models
from Google DeepMind for evaluating the safety of text
prompt input and text output responses against a set of
from Google DeepMind for evaluating the safety of text
prompt input and text output responses against a set of
defined safety policies.
* Fixed issue where ollama pull would leave connections when
* Fixed issue where ollama pull would leave connections when
encountering an error
* ollama rm will now stop a model if it is running prior to
* ollama rm will now stop a model if it is running prior to
deleting it
-------------------------------------------------------------------
Sat Sep 28 03:53:10 UTC 2024 - Alessandro de Oliveira Faria <cabelo@opensuse.org>
- Update to version 0.3.12:
* Llama 3.2: Meta's Llama 3.2 goes small with 1B and 3B
* Llama 3.2: Meta's Llama 3.2 goes small with 1B and 3B
models.
* Qwen 2.5 Coder: The latest series of Code-Specific Qwen
models, with significant improvements in code generation,
* Qwen 2.5 Coder: The latest series of Code-Specific Qwen
models, with significant improvements in code generation,
code reasoning, and code fixing.
* Ollama now supports ARM Windows machines
* Fixed rare issue where Ollama would report a missing .dll
@ -207,23 +269,23 @@ Sun Aug 11 02:40:06 UTC 2024 - Alessandro de Oliveira Faria <cabelo@opensuse.org
- 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
- 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
- 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
- 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
- Batches: generate embeddings for several documents in
one request
- Normalized embeddings: embeddings are now normalized,
- Normalized embeddings: embeddings are now normalized,
improving similarity results
- Truncation: a new truncate parameter that will error if
- Truncation: a new truncate parameter that will error if
set to false
- Metrics: responses include load_duration, total_duration and
- Metrics: responses include load_duration, total_duration and
prompt_eval_count metrics
-------------------------------------------------------------------
@ -234,17 +296,17 @@ Sat Aug 03 09:41:56 UTC 2024 - eyadlorenzo@gmail.com
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
* Fixed issue where /api/generate would respond with an empty
string if provided a context
* Fixed issue where /api/generate would return an incorrect
* 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
* 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 <adrian@suse.de>
@ -309,16 +371,16 @@ Wed Jul 24 14:28:08 UTC 2024 - adrian@suse.de
-------------------------------------------------------------------
Thu Jul 18 13:09:10 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- Fixed issue with shared libraries
- Fixed issue with shared libraries
-------------------------------------------------------------------
Thu Jul 18 12:27:54 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- Added %check section
- Use -v when building
- Use -v when building
- Update to version 0.2.6:
* New models: MathΣtral is a 7B model designed for math
* 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
@ -332,62 +394,62 @@ Sun Jul 14 17:48:36 UTC 2024 - eyadlorenzo@gmail.com
* 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
* 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
* 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
* 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
* 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
* 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
* 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.
* 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
* 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
* New models: CodeGeeX4: A versatile model for AI software
development scenarios, including code completion.
* New models: Gemma 2: Improved output quality and base text
* 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
* 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
* 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 <eyadlorenzo@gmail.com>
- Update to version 0.1.48:
* Fixed issue where Gemma 2 would continuously output when
* 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
* 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 <eyadlorenzo@gmail.com>
@ -422,44 +484,44 @@ Sat Jun 22 10:08:00 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
-------------------------------------------------------------------
Tue Jun 18 12:12:41 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- Added documentation files to .spec
- Added documentation files to .spec
- Update to version 0.1.44:
* Fixed issue where unicode characters such as emojis would not
* 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
* New import.md guide for converting and importing models to
Ollama
* Fixed issue where embedding vectors resulting from
* Fixed issue where embedding vectors resulting from
/api/embeddings would not be accurate
* JSON mode responses will no longer include invalid escape
* JSON mode responses will no longer include invalid escape
characters
* Removing a model will no longer show incorrect File not found
* Removing a model will no longer show incorrect File not found
errors
* Fixed issue where running ollama create would result in an
* 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
* 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
* Qwen 2: a new series of large language models from Alibaba
group
* ollama pull is now faster if it detects a model is already
* 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
for popular model architectures such as Llama, Gemma, Phi and
more.
* Ollama can now be accessed from local apps built with Electron
* 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
* 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
* Fixed issue on Windows 10 and 11 with Intel CPUs with
integrated GPUs where Ollama would encounter an error
-------------------------------------------------------------------
@ -469,12 +531,12 @@ Sat Jun 01 21:12:20 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
* New model: Codestral: Codestral is Mistral AI’
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
* New model: Deepseek V2: A Strong, Economical, and Efficient
Mixture-of-Experts Language Model
* Fixed out of memory and incorrect token issues when running
* 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
* Fixed issue where full-width characters (e.g. Japanese,
Chinese, Russian) were deleted at end of the line when using
ollama run
-------------------------------------------------------------------
@ -483,9 +545,9 @@ Wed May 29 11:38:26 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- 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
* 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,
* 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
@ -493,7 +555,7 @@ Wed May 29 11:38:26 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
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
* Full changelog at
https://github.com/ollama/ollama/releases/tag/v0.1.39
-------------------------------------------------------------------
@ -507,7 +569,7 @@ Thu May 16 19:55:51 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- 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
* 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
@ -532,7 +594,7 @@ Sun May 12 19:05:53 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
Sun May 12 15:20:28 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
- Use obs_scm service instead of the deprecated tar_scm
- Use zstd for vendor tarball compression
- Use zstd for vendor tarball compression
-------------------------------------------------------------------
Sun May 12 01:39:26 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
@ -570,11 +632,11 @@ Sun May 12 01:39:26 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
* New model: CodeGemma 1.1
* New model: StableLM2 12B
* New model: Moondream 2
* Fixed issues with LLaVa models where they would respond
* 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
* Fixed out of memory errors when running large models such as
Llama 3 70B
* Fixed various issues with Nvidia GPU discovery on Linux and
* 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
@ -592,13 +654,13 @@ Sun May 12 01:39:26 UTC 2024 - Eyad Issa <eyadlorenzo@gmail.com>
* 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
* 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
* Fixed out of memory errors when running Mixtral architecture
models
* Aded experimental concurrency features:
~ OLLAMA_NUM_PARALLEL: Handle multiple requests simultaneously
~ OLLAMA_NUM_PARALLEL: Handle multiple requests simultaneously
for a single model
~ OLLAMA_MAX_LOADED_MODELS: Load multiple models simultaneously
Dominique Leuenberger
Git OBS Bridge