GitLab

Still not complete, needs some refinement to our prediction to understand the
discrete GPUs available space so we can see how many layers fit in each one
since we can't split one layer across multiple GPUs we can't treat free space
as one logical block

This reverts commit f5f245cc, reversing
changes made to 94d37fdc.

this change broke gguf v2 which is incorrectly detected as big endian

Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

Co-authored-by: ManniX-ITA <20623405+mann1x@users.noreply.github.com>

- FROM /path/to/{safetensors,pytorch}
- FROM /path/to/fp{16,32}.bin
- FROM model:fp{16,32}

count each layer independently when deciding gpu offloading

Co-authored-by: Michael Yang <mxyng@pm.me>

* select layers based on estimated model memory usage

* always account for scratch vram

* dont load +1 layers

* better estmation for graph alloc

* Update gpu/gpu_darwin.go
Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

* Update llm/llm.go
Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

* Update llm/llm.go

* add overhead for cuda memory

* Update llm/llm.go
Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

* fix build error on linux

* address comments

---------
Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

- remove ggml runner
- automatically pull gguf models when ggml detected
- tell users to update to gguf in the case automatic pull fails
Co-Authored-By: Jeffrey Morgan <jmorganca@gmail.com>

ggufv3 adds support for big endianness, mainly for s390x architecture.
while that's not currently supported for ollama, the change is simple.

loosen version check to be more forward compatible. unless specified,
gguf versions other v1 will be decoded into v2.

---------
Co-authored-by: Michael Yang <mxyng@pm.me>