Tags

Tags give the ability to mark specific points in history as being important

v0.1.33-rc1

de4ded68 · Merge pull request #3923 from ollama/mxyng/mem · Apr 25, 2024
v0.1.32

fb9580df · Merge pull request #3684 from ollama/mxyng/scale-graph · Apr 16, 2024
v0.1.32-rc2

7027f264 · app: gracefully shut down `ollama serve` on windows (#3641) · Apr 14, 2024
v0.1.32-rc1

4c7db6b7 · Merge pull request #3566 from dhiltgen/more_time · Apr 09, 2024
v0.1.31

dc011d16 · Backport MacOS SDK fix from main · Apr 04, 2024
v0.1.30

756c2575 · Merge pull request #3380 from ollama/mxyng/conditional-generate · Mar 28, 2024
v0.1.29

e87c780f · Merge pull request #3149 from ollama/mxyng/fix-memory-leak · Mar 14, 2024
v0.1.28

21347e1e · update llama.cpp submodule to `c29af7e` (#2868) · Mar 01, 2024
v0.1.27

8782dd56 · fix `build_windows.ps1` script to run `go build` with the correct flags · Feb 22, 2024
v0.1.26

2a7553ce · update llama.cpp submodule to `c14f72d` · Feb 21, 2024
v0.1.25

42e77e2a · handle race condition while setting raw mode in windows (#2509) · Feb 14, 2024
v0.1.24

69f392c9 · Merge pull request #2403 from dhiltgen/handle_tmp_cleanup · Feb 07, 2024
v0.1.23

09a6f76f · fix error on `ollama run` with a non-existent model · Feb 01, 2024
v0.1.22

a47d8b25 · Merge pull request #2197 from dhiltgen/remove_rocm_image · Jan 26, 2024
v0.1.21

3ebd6a83 · update submodule to `cd4fddb29f81d6a1f6d51a0c016bc6b486d68def` · Jan 25, 2024
v0.1.20

ab6be852 · revisit memory allocation to account for full kv cache on main gpu · Jan 11, 2024
v0.1.19

34344d80 · clean up cmake `build` directory when cross compiling macOS builds · Jan 09, 2024
v0.1.18

3a9f4471 · only pull gguf model if already exists (#1817) · Jan 05, 2024
v0.1.17

6b5bdfa6 · update runner submodule · Dec 18, 2023
v0.1.16

6ee8c801 · restore model load duration on generate response (#1524) · Dec 14, 2023

Prev
1
2
3
4
Next