Tags
Tags give the ability to mark specific points in history as being important
master-d1f563a
d1f563a7
·
llama : fix Metal KV cache sync (close #1695)
·
Jun 05, 2023
master-ecb217d
ecb217db
·
llama : Metal inference (#1642)
·
Jun 04, 2023
master-dcb2ed4
dcb2ed48
·
OpenCL: Fix duplication of layers in VRAM and RAM, add GPU mul kernel (#1653)
·
Jun 04, 2023
master-136476e
136476e8
·
Fix prompt cache saving and chat-persistent rollover (#1678)
·
Jun 03, 2023
master-ffb06a3
ffb06a34
·
OpenLLaMA 3B support (#1588)
·
May 30, 2023
master-7552ac5
7552ac58
·
ggml : sync cgraph import / export API
·
May 29, 2023
master-2483676
24836760
·
Work around for recalculating logits in cached prompts (Fixes #1585) (#1609)
·
May 29, 2023
master-3b126f6
3b126f65
·
LLAMA_DEBUG adds debug symbols (#1617)
·
May 28, 2023
master-1b78ed2
1b78ed20
·
Only show -ngl option when relevant + other doc/arg handling updates (#1625)
·
May 28, 2023
master-337aea1
337aea11
·
examples : add --alias option to gpt_params to set use friendly model name (#1614)
·
May 28, 2023
master-bb051d9
bb051d97
·
opencl : no need to allocate cl_mem on heap (#1612)
·
May 28, 2023
master-ca74884
ca74884f
·
opencl : use strstr to check if fp16 supported (#1611)
·
May 28, 2023
master-a670464
a6704643
·
ggml : add support for the RISCV architecture (#1616)
·
May 27, 2023
master-0df7d63
0df7d63e
·
Include server in releases + other build system cleanups (#1610)
·
May 27, 2023
master-0ecb1bb
0ecb1bbb
·
[CI] Fix openblas (#1613)
·
May 27, 2023
master-9361803
93618031
·
ggml : add ggml_tensor_overhead()
·
May 27, 2023
master-66874d4
66874d4f
·
Some improvements to loading the session with --prompt-cache (#1550)
·
May 25, 2023
master-1fcdcc2
1fcdcc28
·
cuda : performance optimizations (#1530)
·
May 26, 2023
master-ac7876a
ac7876ac
·
Update CLBlast to 1.6.0 (#1580)
·
May 24, 2023
master-7d87381
7d873811
·
Fix handling of "invalid property" when creating OpenCL command queue (#1565)
·
May 23, 2023
1
…
17
18
19
20
21
22
23
24
25
…
38