mirror of
https://github.com/mudler/LocalAI.git
synced 2024-12-24 06:46:39 +00:00
Drop ggml-based gpt2 and starcoder (supported by llama.cpp) (#1679)
* Drop ggml-based gpt2 and starcoder (supported by llama.cpp) * Update compatibility table
This commit is contained in:
parent
38e4ec0b2a
commit
98ad93d53e
10
Makefile
10
Makefile
@ -145,7 +145,7 @@ ifeq ($(findstring tts,$(GO_TAGS)),tts)
|
||||
OPTIONAL_GRPC+=backend-assets/grpc/piper
|
||||
endif
|
||||
|
||||
ALL_GRPC_BACKENDS=backend-assets/grpc/langchain-huggingface backend-assets/grpc/bert-embeddings backend-assets/grpc/llama backend-assets/grpc/llama-cpp backend-assets/grpc/llama-ggml backend-assets/grpc/gpt4all backend-assets/grpc/dolly backend-assets/grpc/gpt2 backend-assets/grpc/gptj backend-assets/grpc/gptneox backend-assets/grpc/mpt backend-assets/grpc/replit backend-assets/grpc/starcoder backend-assets/grpc/rwkv backend-assets/grpc/whisper $(OPTIONAL_GRPC)
|
||||
ALL_GRPC_BACKENDS=backend-assets/grpc/langchain-huggingface backend-assets/grpc/bert-embeddings backend-assets/grpc/llama backend-assets/grpc/llama-cpp backend-assets/grpc/llama-ggml backend-assets/grpc/gpt4all backend-assets/grpc/dolly backend-assets/grpc/gptj backend-assets/grpc/gptneox backend-assets/grpc/mpt backend-assets/grpc/replit backend-assets/grpc/rwkv backend-assets/grpc/whisper $(OPTIONAL_GRPC)
|
||||
GRPC_BACKENDS?=$(ALL_GRPC_BACKENDS) $(OPTIONAL_GRPC)
|
||||
|
||||
# If empty, then we build all
|
||||
@ -509,10 +509,6 @@ backend-assets/grpc/dolly: backend-assets/grpc sources/go-ggml-transformers/libt
|
||||
CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/sources/go-ggml-transformers LIBRARY_PATH=$(CURDIR)/sources/go-ggml-transformers \
|
||||
$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/dolly ./backend/go/llm/dolly/
|
||||
|
||||
backend-assets/grpc/gpt2: backend-assets/grpc sources/go-ggml-transformers/libtransformers.a
|
||||
CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/sources/go-ggml-transformers LIBRARY_PATH=$(CURDIR)/sources/go-ggml-transformers \
|
||||
$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/gpt2 ./backend/go/llm/gpt2/
|
||||
|
||||
backend-assets/grpc/gptj: backend-assets/grpc sources/go-ggml-transformers/libtransformers.a
|
||||
CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/sources/go-ggml-transformers LIBRARY_PATH=$(CURDIR)/sources/go-ggml-transformers \
|
||||
$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/gptj ./backend/go/llm/gptj/
|
||||
@ -529,10 +525,6 @@ backend-assets/grpc/replit: backend-assets/grpc sources/go-ggml-transformers/lib
|
||||
CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/sources/go-ggml-transformers LIBRARY_PATH=$(CURDIR)/sources/go-ggml-transformers \
|
||||
$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/replit ./backend/go/llm/replit/
|
||||
|
||||
backend-assets/grpc/starcoder: backend-assets/grpc sources/go-ggml-transformers/libtransformers.a
|
||||
CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/sources/go-ggml-transformers LIBRARY_PATH=$(CURDIR)/sources/go-ggml-transformers \
|
||||
$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/starcoder ./backend/go/llm/starcoder/
|
||||
|
||||
backend-assets/grpc/rwkv: backend-assets/grpc sources/go-rwkv/librwkv.a
|
||||
CGO_LDFLAGS="$(CGO_LDFLAGS)" C_INCLUDE_PATH=$(CURDIR)/sources/go-rwkv LIBRARY_PATH=$(CURDIR)/sources/go-rwkv \
|
||||
$(GOCMD) build -ldflags "$(LD_FLAGS)" -tags "$(GO_TAGS)" -o backend-assets/grpc/rwkv ./backend/go/llm/rwkv
|
||||
|
@ -1,23 +0,0 @@
|
||||
package main
|
||||
|
||||
// Note: this is started internally by LocalAI and a server is allocated for each model
|
||||
|
||||
import (
|
||||
"flag"
|
||||
|
||||
transformers "github.com/go-skynet/LocalAI/backend/go/llm/transformers"
|
||||
|
||||
grpc "github.com/go-skynet/LocalAI/pkg/grpc"
|
||||
)
|
||||
|
||||
var (
|
||||
addr = flag.String("addr", "localhost:50051", "the address to connect to")
|
||||
)
|
||||
|
||||
func main() {
|
||||
flag.Parse()
|
||||
|
||||
if err := grpc.StartServer(*addr, &transformers.GPT2{}); err != nil {
|
||||
panic(err)
|
||||
}
|
||||
}
|
@ -1,23 +0,0 @@
|
||||
package main
|
||||
|
||||
// Note: this is started internally by LocalAI and a server is allocated for each model
|
||||
|
||||
import (
|
||||
"flag"
|
||||
|
||||
transformers "github.com/go-skynet/LocalAI/backend/go/llm/transformers"
|
||||
|
||||
grpc "github.com/go-skynet/LocalAI/pkg/grpc"
|
||||
)
|
||||
|
||||
var (
|
||||
addr = flag.String("addr", "localhost:50051", "the address to connect to")
|
||||
)
|
||||
|
||||
func main() {
|
||||
flag.Parse()
|
||||
|
||||
if err := grpc.StartServer(*addr, &transformers.Starcoder{}); err != nil {
|
||||
panic(err)
|
||||
}
|
||||
}
|
@ -16,18 +16,16 @@ LocalAI will attempt to automatically load models which are not explicitly confi
|
||||
|
||||
| Backend and Bindings | Compatible models | Completion/Chat endpoint | Capability | Embeddings support | Token stream support | Acceleration |
|
||||
|----------------------------------------------------------------------------------|-----------------------|--------------------------|---------------------------|-----------------------------------|----------------------|--------------|
|
||||
| [llama.cpp]({{%relref "docs/features/text-generation#llama.cpp" %}}) | Vicuna, Alpaca, LLaMa, Falcon, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | yes | GPT and Functions | yes** | yes | CUDA, openCL, cuBLAS, Metal |
|
||||
| [llama.cpp]({{%relref "docs/features/text-generation#llama.cpp" %}}) | Vicuna, Alpaca, LLaMa, Falcon, Starcoder, GPT-2, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | yes | GPT and Functions | yes** | yes | CUDA, openCL, cuBLAS, Metal |
|
||||
| [gpt4all-llama](https://github.com/nomic-ai/gpt4all) | Vicuna, Alpaca, LLaMa | yes | GPT | no | yes | N/A |
|
||||
| [gpt4all-mpt](https://github.com/nomic-ai/gpt4all) | MPT | yes | GPT | no | yes | N/A |
|
||||
| [gpt4all-j](https://github.com/nomic-ai/gpt4all) | GPT4ALL-J | yes | GPT | no | yes | N/A |
|
||||
| [falcon-ggml](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | Falcon (*) | yes | GPT | no | no | N/A |
|
||||
| [gpt2](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | GPT2, Cerebras | yes | GPT | no | no | N/A |
|
||||
| [dolly](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | Dolly | yes | GPT | no | no | N/A |
|
||||
| [gptj](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | GPTJ | yes | GPT | no | no | N/A |
|
||||
| [mpt](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | MPT | yes | GPT | no | no | N/A |
|
||||
| [replit](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | Replit | yes | GPT | no | no | N/A |
|
||||
| [gptneox](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | GPT NeoX, RedPajama, StableLM | yes | GPT | no | no | N/A |
|
||||
| [starcoder](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp)) | Starcoder | yes | GPT | no | no | N/A|
|
||||
| [bloomz](https://github.com/NouamaneTazi/bloomz.cpp) ([binding](https://github.com/go-skynet/bloomz.cpp)) | Bloom | yes | GPT | no | no | N/A |
|
||||
| [rwkv](https://github.com/saharNooby/rwkv.cpp) ([binding](https://github.com/donomii/go-rwkv.cpp)) | rwkv | yes | GPT | no | yes | N/A |
|
||||
| [bert](https://github.com/skeskinen/bert.cpp) ([binding](https://github.com/go-skynet/go-bert.cpp)) | bert | no | Embeddings only | yes | no | N/A |
|
||||
@ -47,6 +45,7 @@ LocalAI will attempt to automatically load models which are not explicitly confi
|
||||
| [tinydream](https://github.com/symisc/tiny-dream#tiny-dreaman-embedded-header-only-stable-diffusion-inference-c-librarypixlabiotiny-dream) | stablediffusion | no | Image | no | no | N/A |
|
||||
| `coqui` | Coqui | no | Audio generation and Voice cloning | no | no | CPU/CUDA |
|
||||
| `petals` | Various GPTs and quantization formats | yes | GPT | no | no | CPU/CUDA |
|
||||
| `transformers` | Various GPTs and quantization formats | yes | GPT, embeddings | yes | no | CPU/CUDA |
|
||||
|
||||
Note: any backend name listed above can be used in the `backend` field of the model configuration file (See [the advanced section]({{%relref "docs/advanced" %}})).
|
||||
|
||||
|
@ -23,13 +23,11 @@ const (
|
||||
GoLlamaBackend = "llama"
|
||||
LlamaGGML = "llama-ggml"
|
||||
LLamaCPP = "llama-cpp"
|
||||
StarcoderBackend = "starcoder"
|
||||
GPTJBackend = "gptj"
|
||||
DollyBackend = "dolly"
|
||||
MPTBackend = "mpt"
|
||||
GPTNeoXBackend = "gptneox"
|
||||
ReplitBackend = "replit"
|
||||
Gpt2Backend = "gpt2"
|
||||
Gpt4AllLlamaBackend = "gpt4all-llama"
|
||||
Gpt4AllMptBackend = "gpt4all-mpt"
|
||||
Gpt4AllJBackend = "gpt4all-j"
|
||||
@ -55,11 +53,9 @@ var AutoLoadBackends []string = []string{
|
||||
GPTNeoXBackend,
|
||||
BertEmbeddingsBackend,
|
||||
GPTJBackend,
|
||||
Gpt2Backend,
|
||||
DollyBackend,
|
||||
MPTBackend,
|
||||
ReplitBackend,
|
||||
StarcoderBackend,
|
||||
RwkvBackend,
|
||||
WhisperBackend,
|
||||
StableDiffusionBackend,
|
||||
|
Loading…
Reference in New Issue
Block a user