Michael Yang
|
b732beba6a
lint
|
9 miesięcy temu |
Michael Yang
|
5c1912769e
Merge pull request #5473 from ollama/mxyng/environ
|
9 miesięcy temu |
royjhan
|
1b44d873e7
Add Metrics to `api\embed` response (#5709)
|
9 miesięcy temu |
Tibor Schmidt
|
f3d7a481b7
feat: add support for min_p (resolve #1142) (#1825)
|
9 miesięcy temu |
Daniel Hiltgen
|
e12fff8810
Enable windows error dialog for subprocess startup
|
9 miesięcy temu |
Michael Yang
|
e2c3f6b3e2
string
|
10 miesięcy temu |
Michael Yang
|
55cd3ddcca
bool
|
10 miesięcy temu |
Michael Yang
|
35b89b2eab
rfc: dynamic environ lookup
|
10 miesięcy temu |
Daniel Hiltgen
|
a3c20e3f18
Refine error reporting for subprocess crash
|
9 miesięcy temu |
Daniel Hiltgen
|
283948c83b
Adjust windows ROCm discovery
|
9 miesięcy temu |
royjhan
|
b9f5e16c80
Introduce `/api/embed` endpoint supporting batch embedding (#5127)
|
9 miesięcy temu |
Jeffrey Morgan
|
ef98803d63
llm: looser checks for minimum memory (#5677)
|
9 miesięcy temu |
Jeffrey Morgan
|
c4cf8ad559
llm: avoid loading model if system memory is too small (#5637)
|
9 miesięcy temu |
Jeffrey Morgan
|
791650ddef
sched: only error when over-allocating system memory (#5626)
|
9 miesięcy temu |
Daniel Hiltgen
|
22c81f62ec
Remove duplicate merge glitch
|
9 miesięcy temu |
Michael Yang
|
9bbddc37a7
Merge pull request #5126 from ollama/mxyng/messages
|
9 miesięcy temu |
Jeffrey Morgan
|
53da2c6965
llm: remove ambiguous comment when putting upper limit on predictions to avoid infinite generation (#5535)
|
9 miesięcy temu |
Michael Yang
|
ac7a842e55
fix model reloading
|
10 miesięcy temu |
Daniel Hiltgen
|
ccd7785859
Merge pull request #5243 from dhiltgen/modelfile_use_mmap
|
10 miesięcy temu |
Daniel Hiltgen
|
0e982bc1f4
Fix corner cases on tmp cleaner on mac
|
10 miesięcy temu |
Josh Yan
|
33a65e3ba3
error
|
10 miesięcy temu |
Daniel Hiltgen
|
97c9e11768
Switch use_mmap to a pointer type
|
10 miesięcy temu |
Daniel Hiltgen
|
3518aaef33
Merge pull request #4218 from dhiltgen/auto_parallel
|
10 miesięcy temu |
Blake Mizerany
|
cb42e607c5
llm: speed up gguf decoding by a lot (#5246)
|
10 miesięcy temu |
Daniel Hiltgen
|
17b7186cd7
Enable concurrency by default
|
1 rok temu |
Daniel Hiltgen
|
5bf5aeec01
Refine mmap default logic on linux
|
10 miesięcy temu |
Daniel Hiltgen
|
96624aa412
Merge pull request #5072 from dhiltgen/windows_path
|
10 miesięcy temu |
Daniel Hiltgen
|
7784ca33ce
Tighten up memory prediction logging
|
10 miesięcy temu |
Daniel Hiltgen
|
171796791f
Adjust mmap logic for cuda windows for faster model load
|
10 miesięcy temu |
Daniel Hiltgen
|
b2799f111b
Move libraries out of users path
|
10 miesięcy temu |