feat: update go-gpt2 (#359)

Signed-off-by: mudler <mudler@mocaccino.org>

.github/workflows/bump_deps.yaml

@@ -12,8 +12,8 @@ jobs:
           - repository: "go-skynet/go-llama.cpp"
             variable: "GOLLAMA_VERSION"
             branch: "master"
-          - repository: "go-skynet/go-gpt2.cpp"
+          - repository: "go-skynet/go-ggml-transformers.cpp"
-            variable: "GOGPT2_VERSION"
+            variable: "GOGGMLTRANSFORMERS_VERSION"
             branch: "master"
           - repository: "donomii/go-rwkv.cpp"
             variable: "RWKV_VERSION"

.github/workflows/release.yaml

@@ -2,6 +2,9 @@ name: Build and Release
 
 on: push
 
+permissions:
+  contents: write
+
 jobs:
   build-linux:
     strategy:

Makefile

@@ -6,7 +6,7 @@ BINARY_NAME=local-ai
 GOLLAMA_VERSION?=ccf23adfb278c0165d388389a5d60f3fe38e4854
 GPT4ALL_REPO?=https://github.com/nomic-ai/gpt4all
 GPT4ALL_VERSION?=8119ff4df0a99bde44255db2b8c7290b5582ac2b
-GOGPT2_VERSION?=7bff56f0224502c1c9ed6258d2a17e8084628827
+GOGGMLTRANSFORMERS_VERSION?=14fd6c9
 RWKV_REPO?=https://github.com/donomii/go-rwkv.cpp
 RWKV_VERSION?=07166da10cb2a9e8854395a4f210464dcea76e47
 WHISPER_CPP_VERSION?=041be06d5881d3c759cc4ed45d655804361237cd
@@ -29,8 +29,8 @@ WHITE  := $(shell tput -Txterm setaf 7)
 CYAN   := $(shell tput -Txterm setaf 6)
 RESET  := $(shell tput -Txterm sgr0)
 
-C_INCLUDE_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-gpt2:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
+C_INCLUDE_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-ggml-transformers:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
-LIBRARY_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-gpt2:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
+LIBRARY_PATH=$(shell pwd)/go-llama:$(shell pwd)/go-stable-diffusion/:$(shell pwd)/gpt4all/gpt4all-bindings/golang/:$(shell pwd)/go-ggml-transformers:$(shell pwd)/go-rwkv:$(shell pwd)/whisper.cpp:$(shell pwd)/go-bert:$(shell pwd)/bloomz
 
 ifeq ($(BUILD_TYPE),openblas)
 	CGO_LDFLAGS+=-lopenblas
@@ -117,23 +117,23 @@ gpt4all/gpt4all-bindings/golang/libgpt4all.a: gpt4all
 	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ libgpt4all.a
 
 ## CEREBRAS GPT
-go-gpt2: 
+go-ggml-transformers: 
-	git clone --recurse-submodules https://github.com/go-skynet/go-gpt2.cpp go-gpt2
+	git clone --recurse-submodules https://github.com/go-skynet/go-ggml-transformers.cpp go-ggml-transformers
-	cd go-gpt2 && git checkout -b build $(GOGPT2_VERSION) && git submodule update --init --recursive --depth 1
+	cd go-ggml-transformers && git checkout -b build $(GOGPT2_VERSION) && git submodule update --init --recursive --depth 1
 	# This is hackish, but needed as both go-llama and go-gpt4allj have their own version of ggml..
-	@find ./go-gpt2 -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.c" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/ggml_/ggml_gpt2_/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/gpt_print_usage/gpt2_print_usage/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/gpt_params_parse/gpt2_params_parse/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
-	@find ./go-gpt2 -type f -name "*.h" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.h" -exec sed -i'' -e 's/gpt_random_prompt/gpt2_random_prompt/g' {} +
-	@find ./go-gpt2 -type f -name "*.cpp" -exec sed -i'' -e 's/json_/json_gpt2_/g' {} +
+	@find ./go-ggml-transformers -type f -name "*.cpp" -exec sed -i'' -e 's/json_/json_gpt2_/g' {} +
 
-go-gpt2/libgpt2.a: go-gpt2
+go-ggml-transformers/libtransformers.a: go-ggml-transformers
-	$(MAKE) -C go-gpt2 libgpt2.a
+	$(MAKE) -C go-ggml-transformers libtransformers.a
 
 whisper.cpp:
 	git clone https://github.com/ggerganov/whisper.cpp.git
@@ -155,21 +155,21 @@ go-llama/libbinding.a: go-llama
 replace:
 	$(GOCMD) mod edit -replace github.com/go-skynet/go-llama.cpp=$(shell pwd)/go-llama
 	$(GOCMD) mod edit -replace github.com/nomic-ai/gpt4all/gpt4all-bindings/golang=$(shell pwd)/gpt4all/gpt4all-bindings/golang
-	$(GOCMD) mod edit -replace github.com/go-skynet/go-gpt2.cpp=$(shell pwd)/go-gpt2
+	$(GOCMD) mod edit -replace github.com/go-skynet/go-ggml-transformers.cpp=$(shell pwd)/go-ggml-transformers
 	$(GOCMD) mod edit -replace github.com/donomii/go-rwkv.cpp=$(shell pwd)/go-rwkv
 	$(GOCMD) mod edit -replace github.com/ggerganov/whisper.cpp=$(shell pwd)/whisper.cpp
 	$(GOCMD) mod edit -replace github.com/go-skynet/go-bert.cpp=$(shell pwd)/go-bert
 	$(GOCMD) mod edit -replace github.com/go-skynet/bloomz.cpp=$(shell pwd)/bloomz
 	$(GOCMD) mod edit -replace github.com/mudler/go-stable-diffusion=$(shell pwd)/go-stable-diffusion
 
-prepare-sources: go-llama go-gpt2 gpt4all go-rwkv whisper.cpp go-bert bloomz go-stable-diffusion replace
+prepare-sources: go-llama go-ggml-transformers gpt4all go-rwkv whisper.cpp go-bert bloomz go-stable-diffusion replace
 	$(GOCMD) mod download
 
 ## GENERIC
 rebuild: ## Rebuilds the project
 	$(MAKE) -C go-llama clean
 	$(MAKE) -C gpt4all/gpt4all-bindings/golang/ clean
-	$(MAKE) -C go-gpt2 clean
+	$(MAKE) -C go-ggml-transformers clean
 	$(MAKE) -C go-rwkv clean
 	$(MAKE) -C whisper.cpp clean
 	$(MAKE) -C go-stable-diffusion clean
@@ -177,13 +177,13 @@ rebuild: ## Rebuilds the project
 	$(MAKE) -C bloomz clean
 	$(MAKE) build
 
-prepare: prepare-sources gpt4all/gpt4all-bindings/golang/libgpt4all.a $(OPTIONAL_TARGETS) go-llama/libbinding.a go-bert/libgobert.a go-gpt2/libgpt2.a go-rwkv/librwkv.a whisper.cpp/libwhisper.a bloomz/libbloomz.a  ## Prepares for building
+prepare: prepare-sources gpt4all/gpt4all-bindings/golang/libgpt4all.a $(OPTIONAL_TARGETS) go-llama/libbinding.a go-bert/libgobert.a go-ggml-transformers/libtransformers.a go-rwkv/librwkv.a whisper.cpp/libwhisper.a bloomz/libbloomz.a  ## Prepares for building
 
 clean: ## Remove build related file
 	rm -fr ./go-llama
 	rm -rf ./gpt4all
 	rm -rf ./go-stable-diffusion
-	rm -rf ./go-gpt2
+	rm -rf ./go-ggml-transformers
 	rm -rf ./go-rwkv
 	rm -rf ./go-bert
 	rm -rf ./bloomz
@@ -213,7 +213,7 @@ run: prepare ## run local-ai
 test-models/testmodel:
 	mkdir test-models
 	mkdir test-dir
-	wget https://huggingface.co/concedo/cerebras-111M-ggml/resolve/main/cerberas-111m-q4_0.bin -O test-models/testmodel
+	wget https://huggingface.co/nnakasato/ggml-model-test/resolve/main/ggml-model-q4.bin -O test-models/testmodel
 	wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin -O test-models/whisper-en
 	wget https://huggingface.co/skeskinen/ggml/resolve/main/all-MiniLM-L6-v2/ggml-model-q4_0.bin -O test-models/bert
 	wget https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav -O test-dir/audio.wav

README.md

@@ -129,13 +129,13 @@ Depending on the model you are attempting to run might need more RAM or CPU reso
 | [gpt4all-llama](https://github.com/nomic-ai/gpt4all)      | Vicuna, Alpaca, LLaMa | yes                      | no                        | no                                | yes                  |
 | [gpt4all-mpt](https://github.com/nomic-ai/gpt4all)          | MPT                   | yes                      | no                        | no                                | yes                  |
 | [gpt4all-j](https://github.com/nomic-ai/gpt4all)           | GPT4ALL-J             | yes                      | no                        | no                                | yes                  |
-| [gpt2](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))             | GPT/NeoX, Cerebras    | yes                      | no                        | no                                | no                   |
+| [gpt2](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))             | GPT/NeoX, Cerebras    | yes                      | no                        | no                                | no                   |
-| [dolly](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))            | Dolly                 | yes                      | no                        | no                                | no                   |
+| [dolly](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))            | Dolly                 | yes                      | no                        | no                                | no                   |
-| [redpajama](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | RedPajama             | yes                      | no                        | no                                | no                   |
+| [gptj](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | GPTJ             | yes                      | no                        | no                                | no                   |
-| [stableLM](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))         | StableLM GPT/NeoX     | yes                      | no                        | no                                | no                   |
+| [mpt](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))         | MPT     | yes                      | no                        | no                                | no                   |
-| [replit](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | Replit             | yes                      | no                        | no                                | no                   |
+| [replit](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | Replit             | yes                      | no                        | no                                | no                   |
-| [gptneox](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | GPT NeoX             | yes                      | no                        | no                                | no                   |
+| [gptneox](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | GPT NeoX, RedPajama, StableLM             | yes                      | no                        | no                                | no                   |
-| [starcoder](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-gpt2.cpp))        | Starcoder             | yes                      | no                        | no                                | no                   |
+| [starcoder](https://github.com/ggerganov/ggml) ([binding](https://github.com/go-skynet/go-ggml-transformers.cpp))        | Starcoder             | yes                      | no                        | no                                | no                   |
 | [bloomz](https://github.com/NouamaneTazi/bloomz.cpp) ([binding](https://github.com/go-skynet/bloomz.cpp))       | Bloom                 | yes                      | no                        | no                                | no                   |
 | [rwkv](https://github.com/saharNooby/rwkv.cpp) ([binding](https://github.com/donomii/go-rw))       | rwkv                 | yes                      | no                        | no                                | yes                   |
 | [bert](https://github.com/skeskinen/bert.cpp) ([binding](https://github.com/go-skynet/go-bert.cpp) | bert                  | no                       | no                  | yes                               | no                   |    
@@ -1045,7 +1045,7 @@ MIT
 
 - [go-skynet/go-llama.cpp](https://github.com/go-skynet/go-llama.cpp)
 - [go-skynet/go-gpt4all-j.cpp](https://github.com/go-skynet/go-gpt4all-j.cpp)
-- [go-skynet/go-gpt2.cpp](https://github.com/go-skynet/go-gpt2.cpp)
+- [go-skynet/go-ggml-transformers.cpp](https://github.com/go-skynet/go-ggml-transformers.cpp)
 - [go-skynet/go-bert.cpp](https://github.com/go-skynet/go-bert.cpp)
 - [donomii/go-rwkv.cpp](https://github.com/donomii/go-rwkv.cpp)
 

api/prediction.go

@@ -11,7 +11,7 @@ import (
 	"github.com/go-skynet/LocalAI/pkg/stablediffusion"
 	"github.com/go-skynet/bloomz.cpp"
 	bert "github.com/go-skynet/go-bert.cpp"
-	gpt2 "github.com/go-skynet/go-gpt2.cpp"
+	transformers "github.com/go-skynet/go-ggml-transformers.cpp"
 	llama "github.com/go-skynet/go-llama.cpp"
 	gpt4all "github.com/nomic-ai/gpt4all/gpt4all-bindings/golang"
 )
@@ -243,23 +243,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 
 			return response, nil
 		}
-	case *gpt2.GPTNeoX:
+	case *transformers.GPTNeoX:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -267,23 +267,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.Replit:
+	case *transformers.Replit:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -291,23 +291,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.Starcoder:
+	case *transformers.Starcoder:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -315,23 +315,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.RedPajama:
+	case *transformers.MPT:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -359,23 +359,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.StableLM:
+	case *transformers.GPTJ:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -383,23 +383,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.Dolly:
+	case *transformers.Dolly:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(
@@ -407,23 +407,23 @@ func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback
 				predictOptions...,
 			)
 		}
-	case *gpt2.GPT2:
+	case *transformers.GPT2:
 		fn = func() (string, error) {
 			// Generate the prediction using the language model
-			predictOptions := []gpt2.PredictOption{
+			predictOptions := []transformers.PredictOption{
-				gpt2.SetTemperature(c.Temperature),
+				transformers.SetTemperature(c.Temperature),
-				gpt2.SetTopP(c.TopP),
+				transformers.SetTopP(c.TopP),
-				gpt2.SetTopK(c.TopK),
+				transformers.SetTopK(c.TopK),
-				gpt2.SetTokens(c.Maxtokens),
+				transformers.SetTokens(c.Maxtokens),
-				gpt2.SetThreads(c.Threads),
+				transformers.SetThreads(c.Threads),
 			}
 
 			if c.Batch != 0 {
-				predictOptions = append(predictOptions, gpt2.SetBatch(c.Batch))
+				predictOptions = append(predictOptions, transformers.SetBatch(c.Batch))
 			}
 
 			if c.Seed != 0 {
-				predictOptions = append(predictOptions, gpt2.SetSeed(c.Seed))
+				predictOptions = append(predictOptions, transformers.SetSeed(c.Seed))
 			}
 
 			return model.Predict(

go.mod

@@ -8,7 +8,7 @@ require (
 	github.com/go-audio/wav v1.1.0
 	github.com/go-skynet/bloomz.cpp v0.0.0-20230510223001-e9366e82abdf
 	github.com/go-skynet/go-bert.cpp v0.0.0-20230516063724-cea1ed76a7f4
-	github.com/go-skynet/go-gpt2.cpp v0.0.0-20230512145559-7bff56f02245
+	github.com/go-skynet/go-ggml-transformers.cpp v0.0.0-20230523150735-8bfcb3ea6127
 	github.com/go-skynet/go-llama.cpp v0.0.0-20230520155239-ccf23adfb278
 	github.com/gofiber/fiber/v2 v2.46.0
 	github.com/google/uuid v1.3.0
@@ -41,6 +41,7 @@ require (
 	github.com/go-openapi/jsonreference v0.19.6 // indirect
 	github.com/go-openapi/spec v0.20.4 // indirect
 	github.com/go-openapi/swag v0.19.15 // indirect
+	github.com/go-skynet/go-gpt2.cpp v0.0.0-20230523153133-3eb3a32c0874 // indirect
 	github.com/go-task/slim-sprig v0.0.0-20230315185526-52ccab3ef572 // indirect
 	github.com/google/go-cmp v0.5.9 // indirect
 	github.com/google/pprof v0.0.0-20210407192527-94a9f03dee38 // indirect

go.sum

@@ -16,12 +16,6 @@ github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ3
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
-github.com/donomii/go-rwkv.cpp v0.0.0-20230515123100-6fdd0c338e56 h1:s8/MZdicstKi5fn9D9mKGIQ/q6IWCYCk/BM68i8v51w=
-github.com/donomii/go-rwkv.cpp v0.0.0-20230515123100-6fdd0c338e56/go.mod h1:gWy7FIWioqYmYxkaoFyBnaKApeZVrUkHhv9EV9pz4dM=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230515153606-95b02d76b04d h1:uxKTbiRnplE2SubchneSf4NChtxLJtOy9VdHnQMT0d0=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230515153606-95b02d76b04d/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520170006-429b9785c080 h1:W3itqKpRX9FhheKiAxdmuOBy/mjDfMf2G1vcuFIYqZc=
-github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520170006-429b9785c080/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
 github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520182345-041be06d5881 h1:dafqVivljYk51VLFnnpTXJnfWDe637EobWZ1l8PyEf8=
 github.com/ggerganov/whisper.cpp/bindings/go v0.0.0-20230520182345-041be06d5881/go.mod h1:QIjZ9OktHFG7p+/m3sMvrAJKKdWrr1fZIK0rM6HZlyo=
 github.com/go-audio/audio v1.0.0 h1:zS9vebldgbQqktK4H0lUqWrG8P0NxCJVqcj7ZpNnwd4=
@@ -42,21 +36,11 @@ github.com/go-openapi/spec v0.20.4/go.mod h1:faYFR1CvsJZ0mNsmsphTMSoRrNV3TEDoAM7
 github.com/go-openapi/swag v0.19.5/go.mod h1:POnQmlKehdgb5mhVOsnJFsivZCEZ/vjK9gh66Z9tfKk=
 github.com/go-openapi/swag v0.19.15 h1:D2NRCBzS9/pEY3gP9Nl8aDqGUcPFrwG2p+CNFrLyrCM=
 github.com/go-openapi/swag v0.19.15/go.mod h1:QYRuS/SOXUCsnplDa677K7+DxSOj6IPNl/eQntq43wQ=
-github.com/go-skynet/bloomz.cpp v0.0.0-20230510223001-e9366e82abdf h1:VJfSn8hIDE+K5+h38M3iAyFXrxpRExMKRdTk33UDxsw=
+github.com/go-skynet/go-gpt2.cpp v0.0.0-20230523153133-3eb3a32c0874 h1:/6QWh2oarU7iPSpXj/3bLlkKptyxjKTRrNtGUrh8vhI=
-github.com/go-skynet/bloomz.cpp v0.0.0-20230510223001-e9366e82abdf/go.mod h1:wc0fJ9V04yiYTfgKvE5RUUSRQ5Kzi0Bo4I+U3nNOUuA=
+github.com/go-skynet/go-gpt2.cpp v0.0.0-20230523153133-3eb3a32c0874/go.mod h1:1Wj/xbkMfwQSOrhNYK178IzqQHstZbRfhx4s8p1M5VM=
-github.com/go-skynet/go-bert.cpp v0.0.0-20230516063724-cea1ed76a7f4 h1:+3KPDf4Wv1VHOkzAfZnlj9qakLSYggTpm80AswhD/FU=
-github.com/go-skynet/go-bert.cpp v0.0.0-20230516063724-cea1ed76a7f4/go.mod h1:VY0s5KoAI2jRCvQXKuDeEEe8KG7VaWifSNJSk+E1KtY=
-github.com/go-skynet/go-gpt2.cpp v0.0.0-20230512145559-7bff56f02245 h1:IcfYY5uH0DdDXEJKJ8bq0WZCd9guPPd3xllaWNy8LOk=
-github.com/go-skynet/go-gpt2.cpp v0.0.0-20230512145559-7bff56f02245/go.mod h1:1Wj/xbkMfwQSOrhNYK178IzqQHstZbRfhx4s8p1M5VM=
-github.com/go-skynet/go-llama.cpp v0.0.0-20230520082618-a298043ef5f1 h1:i0oM2MERUgMIRmjOcv22TDQULxbmY8o9rZKLKKyWXLo=
-github.com/go-skynet/go-llama.cpp v0.0.0-20230520082618-a298043ef5f1/go.mod h1:oA0r4BW8ndyjTMGi1tulsNd7sdg3Ql8MaVFuT1zF6ws=
-github.com/go-skynet/go-llama.cpp v0.0.0-20230520155239-ccf23adfb278 h1:st4ow9JKy3UuhkwutrbWof2vMFU/YxwBCLYZ1IxJ2Po=
-github.com/go-skynet/go-llama.cpp v0.0.0-20230520155239-ccf23adfb278/go.mod h1:oA0r4BW8ndyjTMGi1tulsNd7sdg3Ql8MaVFuT1zF6ws=
 github.com/go-task/slim-sprig v0.0.0-20230315185526-52ccab3ef572 h1:tfuBGBXKqDEevZMzYi5KSi8KkcZtzBcTgAUUtapy0OI=
 github.com/go-task/slim-sprig v0.0.0-20230315185526-52ccab3ef572/go.mod h1:9Pwr4B2jHnOSGXyyzV8ROjYa2ojvAY6HCGYYfMoC3Ls=
 github.com/godbus/dbus/v5 v5.0.4/go.mod h1:xhWf0FNVPg57R7Z0UbKHbJfkEywrmjJnf7w5xrFpKfA=
-github.com/gofiber/fiber/v2 v2.45.0 h1:p4RpkJT9GAW6parBSbcNFH2ApnAuW3OzaQzbOCoDu+s=
-github.com/gofiber/fiber/v2 v2.45.0/go.mod h1:DNl0/c37WLe0g92U6lx1VMQuxGUQY5V7EIaVoEsUffc=
 github.com/gofiber/fiber/v2 v2.46.0 h1:wkkWotblsGVlLjXj2dpgKQAYHtXumsK/HyFugQM68Ns=
 github.com/gofiber/fiber/v2 v2.46.0/go.mod h1:DNl0/c37WLe0g92U6lx1VMQuxGUQY5V7EIaVoEsUffc=
 github.com/golang/protobuf v1.5.3 h1:KhyjKVUg7Usr/dYsdSqoFveMYd5ko72D+zANwlG1mmg=
@@ -95,20 +79,8 @@ github.com/mattn/go-isatty v0.0.18 h1:DOKFKCQ7FNG2L1rbrmstDN4QVRdS89Nkh85u68Uwp9
 github.com/mattn/go-isatty v0.0.18/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
 github.com/mattn/go-runewidth v0.0.14 h1:+xnbZSEeDbOIg5/mE6JF0w6n9duR1l3/WmbinWVwUuU=
 github.com/mattn/go-runewidth v0.0.14/go.mod h1:Jdepj2loyihRzMpdS35Xk/zdY8IAYHsh153qUoGf23w=
-github.com/mudler/go-stable-diffusion v0.0.0-20230516152536-c0748eca3642 h1:KTkh3lOUsGqQyP4v+oa38sPFdrZtNnM4HaxTb3epdYs=
-github.com/mudler/go-stable-diffusion v0.0.0-20230516152536-c0748eca3642/go.mod h1:8ufRkpz/S/9ahkaxzZ5i4WMgO9w4InEhuRoT7vK5Rnw=
 github.com/niemeyer/pretty v0.0.0-20200227124842-a10e7caefd8e h1:fD57ERR4JtEqsWbfPhv4DMiApHyliiK5xCTNVSPiaAs=
 github.com/niemeyer/pretty v0.0.0-20200227124842-a10e7caefd8e/go.mod h1:zD1mROLANZcx1PVRCS0qkT7pwLkGfwJo4zjcN/Tysno=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230519014017-914519e772fd h1:kMnZASxCNc8GsPuAV94tltEsfT6T+esuB+rgzdjwFVM=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230519014017-914519e772fd/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230521011615-aba1147a2253 h1:7udNpoHYOBktcpCEe8aDaPJ0LyzyRhVjpzAGFjPxPkY=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230521011615-aba1147a2253/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522155256-c8c95ab46f92 h1:brOLJSsTLnFK2vUVi7MaVdxAEhHkOsoboR0vR5WW1HU=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522155256-c8c95ab46f92/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522200803-5ca8767c81a2 h1:3368tGU1ooRSPw0zMvXqv9wLMxS82LzEkVSuo8DWZBI=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522200803-5ca8767c81a2/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522220313-2ce22208a3dd h1:is/rE0YD8oEWcX3fQ+VxoS3fD0LqFEmTxh8XZegYYsA=
-github.com/nomic-ai/gpt4all/gpt4all-bindings/golang v0.0.0-20230522220313-2ce22208a3dd/go.mod h1:4T3CHXyrt+7FQHXaxULZfPjHbD8/99WuDDJa0YVZARI=
 github.com/onsi/ginkgo/v2 v2.9.5 h1:+6Hr4uxzP4XIUyAkg61dWBw8lb/gc4/X5luuxN/EC+Q=
 github.com/onsi/ginkgo/v2 v2.9.5/go.mod h1:tvAoo1QUJwNEU2ITftXTpR7R1RbCzoZUOs3RonqW57k=
 github.com/onsi/gomega v1.27.7 h1:fVih9JD6ogIiHUN6ePK7HJidyEDpWGVB5mzM7cWNXoU=

pkg/model/initializers.go

@@ -10,7 +10,7 @@ import (
 	"github.com/go-skynet/LocalAI/pkg/stablediffusion"
 	bloomz "github.com/go-skynet/bloomz.cpp"
 	bert "github.com/go-skynet/go-bert.cpp"
-	gpt2 "github.com/go-skynet/go-gpt2.cpp"
+	transformers "github.com/go-skynet/go-ggml-transformers.cpp"
 	llama "github.com/go-skynet/go-llama.cpp"
 	"github.com/hashicorp/go-multierror"
 	gpt4all "github.com/nomic-ai/gpt4all/gpt4all-bindings/golang"
@@ -23,9 +23,9 @@ const (
 	LlamaBackend           = "llama"
 	BloomzBackend          = "bloomz"
 	StarcoderBackend       = "starcoder"
-	StableLMBackend        = "stablelm"
+	GPTJBackend            = "gptj"
 	DollyBackend           = "dolly"
-	RedPajamaBackend       = "redpajama"
+	MPTBackend             = "mpt"
 	GPTNeoXBackend         = "gptneox"
 	ReplitBackend          = "replit"
 	Gpt2Backend            = "gpt2"
@@ -43,41 +43,41 @@ var backends []string = []string{
 	Gpt4AllLlamaBackend,
 	Gpt4AllMptBackend,
 	Gpt4AllJBackend,
-	Gpt2Backend,
-	WhisperBackend,
 	RwkvBackend,
-	BloomzBackend,
-	StableLMBackend,
-	DollyBackend,
-	RedPajamaBackend,
-	ReplitBackend,
 	GPTNeoXBackend,
+	WhisperBackend,
 	BertEmbeddingsBackend,
+	GPTJBackend,
+	Gpt2Backend,
+	DollyBackend,
+	MPTBackend,
+	ReplitBackend,
 	StarcoderBackend,
+	BloomzBackend,
 }
 
 var starCoder = func(modelFile string) (interface{}, error) {
-	return gpt2.NewStarcoder(modelFile)
+	return transformers.NewStarcoder(modelFile)
 }
 
-var redPajama = func(modelFile string) (interface{}, error) {
+var mpt = func(modelFile string) (interface{}, error) {
-	return gpt2.NewRedPajama(modelFile)
+	return transformers.NewMPT(modelFile)
 }
 
 var dolly = func(modelFile string) (interface{}, error) {
-	return gpt2.NewDolly(modelFile)
+	return transformers.NewDolly(modelFile)
 }
 
 var gptNeoX = func(modelFile string) (interface{}, error) {
-	return gpt2.NewGPTNeoX(modelFile)
+	return transformers.NewGPTNeoX(modelFile)
 }
 
 var replit = func(modelFile string) (interface{}, error) {
-	return gpt2.NewReplit(modelFile)
+	return transformers.NewReplit(modelFile)
 }
 
-var stableLM = func(modelFile string) (interface{}, error) {
+var gptJ = func(modelFile string) (interface{}, error) {
-	return gpt2.NewStableLM(modelFile)
+	return transformers.NewGPTJ(modelFile)
 }
 
 var bertEmbeddings = func(modelFile string) (interface{}, error) {
@@ -87,8 +87,9 @@ var bertEmbeddings = func(modelFile string) (interface{}, error) {
 var bloomzLM = func(modelFile string) (interface{}, error) {
 	return bloomz.New(modelFile)
 }
-var gpt2LM = func(modelFile string) (interface{}, error) {
+
-	return gpt2.New(modelFile)
+var transformersLM = func(modelFile string) (interface{}, error) {
+	return transformers.New(modelFile)
 }
 
 var stableDiffusion = func(assetDir string) (interface{}, error) {
@@ -130,14 +131,14 @@ func (ml *ModelLoader) BackendLoader(backendString string, modelFile string, lla
 		return ml.LoadModel(modelFile, llamaLM(llamaOpts...))
 	case BloomzBackend:
 		return ml.LoadModel(modelFile, bloomzLM)
-	case StableLMBackend:
+	case GPTJBackend:
-		return ml.LoadModel(modelFile, stableLM)
+		return ml.LoadModel(modelFile, gptJ)
 	case DollyBackend:
 		return ml.LoadModel(modelFile, dolly)
-	case RedPajamaBackend:
+	case MPTBackend:
-		return ml.LoadModel(modelFile, redPajama)
+		return ml.LoadModel(modelFile, mpt)
 	case Gpt2Backend:
-		return ml.LoadModel(modelFile, gpt2LM)
+		return ml.LoadModel(modelFile, transformersLM)
 	case GPTNeoXBackend:
 		return ml.LoadModel(modelFile, gptNeoX)
 	case ReplitBackend:

tests/models_fixtures/embeddings.yaml

@@ -1,6 +1,5 @@
 name: text-embedding-ada-002
 parameters:
   model: bert
-threads: 14
 backend: bert-embeddings
 embeddings: true

tests/models_fixtures/rwkv.yaml

@@ -6,7 +6,6 @@ parameters:
   max_tokens: 100
   top_p: 0.8
 context_size: 1024
-threads: 14
 backend: "rwkv"
 cutwords:
 - "Bob:.*"