ollama

Commit Graph

Author	SHA1	Message	Date
Parth Sareen	108fe02165	sample: make mutations in transforms explicit (#9743 ) * updated minP to use early exit making use of sorted tokens	1 year ago
Daniel Hiltgen	50b5962042	Add support for ROCm gfx1151 (#9773 )	1 year ago
Louis Beaumont	e27e4a3c1b	readme: add screenpipe to community integrations (#9786 )	1 year ago
zeo	088514bbd4	readme: add Ellama to list of community integrations (#9800 )	1 year ago
Patrick Devine	2c8b484643	fix: correctly save in interactive mode (#9788 ) This fixes the case where a FROM line in previous modelfile points to a file which may/may not be present in a different ollama instance. We shouldn't be relying on the filename though and instead just check if the FROM line was instead a valid model name and point to that instead.	1 year ago
Blake Mizerany	8294676150	server/internal/client/ollama: set User-Agent for registry client (#9775 ) This sets the agent header in DefaultRegistry to include the version of the client, OS, and architecture in the previous format, with a minor twist. Note: The version is obtained from the build info, instead of the version in version.Version, which should not longer be necessary, but we can remove in a future commit. Using the build info is more accurate and also provides extra build information if the build is not tagged, and if it is "dirty". Previously, the version was just "0.0.0" with no other helpful information. The ollama.com registry and others handle this swimmingly.	1 year ago
Patrick Devine	ef378ad673	gemma3 quantization (#9776 )	1 year ago
Daniel Hiltgen	2d2247e59e	Align versions for local builds (#9635 ) Darwin was using a different pattern for the version string than linux or windows.	1 year ago
Jesse Gross	7bf793a600	gemma3: Allow multiple image in a single input Previously processing multiple images in a batch would trigger segfaults so sending images together was disabled as a way to mitigate this. The trigger was processing one image on the CPU and one on the GPU. This can no longer happen: - The vision encoder is now on the GPU so both images would be processed on the GPU. - We require images to be fully contained in a batch and each image including its special tokens is over half the batch size. As a result, we will never get two images in the same batch. Fixes #9731	1 year ago
Jesse Gross	282bfaaa95	ollamarunner: Use a separate context per multimodal input Currently there is a single context per sequence, shared all by all multimodal inputs. Since we build a vision encoder graph per image, with a large number of inputs we can eventually hit the maximum number of graph nodes per context. This changes to use a separate context for each image, ensuring that available resource limits are consistent.	1 year ago
Jesse Gross	9679f40146	ml: Allow models to constrain inputs to a single batch Models may require that a set of inputs all be processed as part of the same batch. For example, if an image has multiple patches with fully connected attention between them, we should not split the batch in the middle of an image. Fixes #9697	1 year ago
Bruce MacDonald	3892c3a703	llm: remove internal subprocess req and resp types (#9324 ) This commit refactors the LLM subsystem by removing internal subprocess request and response types. It consolidates duplicate type definitions across the codebase, moving them to centralized locations. The change also standardizes interfaces between components, simplifies the ServerStatusResp struct, and moves the ParseDurationMs function to a common package. This cleanup reduces code duplication between different runner implementations (llamarunner and ollamarunner).	1 year ago
Blake Mizerany	4e320b8b90	server/internal/chunks: remove chunks package (#9755 )	1 year ago
Blake Mizerany	eb2b22b042	server/internal/client: use chunksums for concurrent blob verification (#9746 ) Replace large-chunk blob downloads with parallel small-chunk verification to solve timeout and performance issues. Registry users experienced progressively slowing download speeds as large-chunk transfers aged, often timing out completely. The previous approach downloaded blobs in a few large chunks but required a separate, single-threaded pass to read the entire blob back from disk for verification after download completion. This change uses the new chunksums API to fetch many smaller chunk+digest pairs, allowing concurrent downloads and immediate verification as each chunk arrives. Chunks are written directly to their final positions, eliminating the entire separate verification pass. The result is more reliable downloads that maintain speed throughout the transfer process and significantly faster overall completion, especially over unstable connections or with large blobs.	1 year ago
Michael Yang	4ea4d2b189	Merge pull request #9703 from ollama/mxyng/gemma3-memory count gemma3 vision tensors	1 year ago
Michael Yang	8d76fa23ef	count non-repeating vision layers	1 year ago
Bradley Erickson	74b44fdf8f	docs: Add OLLAMA_ORIGINS for browser extension support (#9643 )	1 year ago
Michael Yang	65b88c544f	fix divide by zero	1 year ago
Michael Yang	a422ba39c9	roughly count gemma3 graph the largest operation is by far (q @ k) so just count that for simplicity	1 year ago
Michael Yang	d2ec22371e	count all vision tensors	1 year ago
Michael Yang	033cec232a	count gemma3 vision tensors	1 year ago
Michael Yang	543240fb5f	Merge pull request #9741 from ollama/mxyng/visionless fix: error if image requested without vision model	1 year ago
Patrick Devine	4bed739259	add verbose mode to the show command (#9640 ) Add metadata and tensor information to the show command to be able to see more information about a model. This outputs the same data as shown on the model details page on ollama.com	1 year ago
Patrick Devine	80c7ce381b	fix: change default context size for gemma3 (#9744 )	1 year ago
Michael Yang	ccfd41c4f0	Merge pull request #9742 from ollama/mxyng/engine-error-embeddings fix: error on models that don't support embeddings	1 year ago
Michael Yang	3e102b7dad	Update model/model.go Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>	1 year ago
Michael Yang	ec46f3286c	engine: error on embeddings; not currently implemented	1 year ago
Michael Yang	5e2e0b46b1	fix: error if image requested without vision model	1 year ago
Michael Yang	45a13b1dec	Merge pull request #9688 from Shane-XB-Qian/debug_mistype_lld ollama-debug.c: correct mistype	1 year ago
Parth Sareen	5c0b663969	sample: separate softmax and temperature transforms (#9732 )	1 year ago
shane.xb.qian	30d7a59ba8	ollama-debug.c: change 'ld' to 'PRIi64' * macOS has different definition per info from @mxyng	1 year ago
ParthSareen	4aeb67ef4c	sample: do all sorting in topK	1 year ago
ParthSareen	3ba91634c1	sample: simplify top_k=0 sorting	1 year ago
ParthSareen	1b7433b71e	sample: use container/heap for top_k	1 year ago
Bruce MacDonald	a70820daa0	models/gemma3: remove final logit softcap (#9692 ) Softcap isn't in the whitepaper/implementation for the language model so we should remove it. There is no discernible difference in output with it removed.	1 year ago
Shane-XB-Qian	6b45b1d6b4	cli: adding support ctrl-n/p like general cli (#9136 ) Signed-off-by: shane.xb.qian <shane.qian@foxmail.com>	1 year ago
shane.xb.qian	85ab552028	ollama-debug.c: correct mistype Signed-off-by: shane.xb.qian <shane.qian@foxmail.com>	1 year ago
frob	b3af953a55	cli: don't exit for invalid model during /load. (#9576 ) Co-authored-by: Richard Lyons <frob@cloudstaff.com>	1 year ago
Michael	ad4e0bf3be	Adding Gemma 3 to readme (#9671 )	1 year ago
Michael Yang	aee28501b5	Merge pull request #9661 from ollama/gemma engine: add gemma support	1 year ago
jmorganca	83f0ec8269	all: address linter errors	1 year ago
jmorganca	c6b6938b3a	kvcache: fix tests by adding AvgPool2D stub	1 year ago
jmorganca	fb4664fcec	model: add more spm tokenizer tests	1 year ago
jmorganca	20e3593863	model: validate left and right pairs before merging them	1 year ago
Michael Yang	63a394068c	use 2d pooling	1 year ago
Daniel Hiltgen	ab39e08eb9	llm: auto detect models that require Ollama Engine (#1 )	1 year ago
jmorganca	11bfa62796	add trailing \n\n after <end_of_image> to match reference implementation	1 year ago
jmorganca	f63e62e546	reduce kernel size, add TODO for loading from config	1 year ago
jmorganca	65b0f329d1	Revert "Allow models to force a new batch" This reverts commit c7eae586b899083acebcd9b3847b89ea78c2850c.	1 year ago
Jesse Gross	06007c0a18	Allow models to force a new batch This is useful for a few things: - Work around bugs, such as having 2 images in one batch - Keep the image in a single batch for fully connected attention - Improve performance by not evaluating embeddings multiple times	1 year ago

1 2 3 4 5 ...

4076 Commits (108fe021657ea1f7a012299f424d673e86512df2) All Branches Search

4076 Commits (108fe021657ea1f7a012299f424d673e86512df2)

All Branches