LocalAI/core/http/endpoints/openai/completion.go
mintyleaf 96306a39a0
Some checks are pending
Explorer deployment / build-linux (push) Waiting to run
GPU tests / ubuntu-latest (1.21.x) (push) Waiting to run
generate and publish GRPC docker caches / generate_caches (ubuntu:22.04, linux/amd64,linux/arm64, ubuntu-latest) (push) Waiting to run
generate and publish intel docker caches / generate_caches (intel/oneapi-basekit:2025.0.0-0-devel-ubuntu22.04, linux/amd64, ubuntu-latest) (push) Waiting to run
build container images / hipblas-jobs (-aio-gpu-hipblas, rocm/dev-ubuntu-22.04:6.1, hipblas, true, ubuntu:22.04, extras, latest-gpu-hipblas, latest-aio-gpu-hipblas, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -hipblas-ffmpeg) (push) Waiting to run
build container images / hipblas-jobs (rocm/dev-ubuntu-22.04:6.1, hipblas, false, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -hipblas-core) (push) Waiting to run
build container images / hipblas-jobs (rocm/dev-ubuntu-22.04:6.1, hipblas, false, ubuntu:22.04, extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -hipblas) (push) Waiting to run
build container images / hipblas-jobs (rocm/dev-ubuntu-22.04:6.1, hipblas, true, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -hipblas-ffmpeg-core) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-intel-f16, quay.io/go-skynet/intel-oneapi-base:latest, sycl_f16, true, ubuntu:22.04, extras, latest-gpu-intel-f16, latest-aio-gpu-intel-f16, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -sycl-f16-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-intel-f32, quay.io/go-skynet/intel-oneapi-base:latest, sycl_f32, true, ubuntu:22.04, extras, latest-gpu-intel-f32, latest-aio-gpu-intel-f32, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -sycl-f32-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-nvidia-cuda-11, ubuntu:22.04, cublas, 11, 7, true, extras, latest-gpu-nvidia-cuda-11, latest-aio-gpu-nvidia-cuda-11, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -cublas-cuda11-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (-aio-gpu-nvidia-cuda-12, ubuntu:22.04, cublas, 12, 0, true, extras, latest-gpu-nvidia-cuda-12, latest-aio-gpu-nvidia-cuda-12, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -cublas-cuda12-ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f16, false, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f16-core) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f16, true, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f16-ffmpeg-core) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f32, false, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f32-core) (push) Waiting to run
build container images / self-hosted-jobs (quay.io/go-skynet/intel-oneapi-base:latest, sycl_f32, true, ubuntu:22.04, core, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -sycl-f32-ffmpeg-core) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, , , extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, ) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, , true, extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, auto, -ffmpeg) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, cublas, 11, 7, , extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda11) (push) Waiting to run
build container images / self-hosted-jobs (ubuntu:22.04, cublas, 12, 0, , extras, --jobs=3 --output-sync=target, linux/amd64, arc-runner-set, false, -cublas-cuda12) (push) Waiting to run
build container images / core-image-build (-aio-cpu, ubuntu:22.04, , true, core, latest-cpu, latest-aio-cpu, --jobs=4 --output-sync=target, linux/amd64,linux/arm64, arc-runner-set, false, auto, -ffmpeg-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 11, 7, , core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, false, -cublas-cuda11-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 11, 7, true, core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, false, -cublas-cuda11-ffmpeg-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 12, 0, , core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, false, -cublas-cuda12-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, cublas, 12, 0, true, core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, false, -cublas-cuda12-ffmpeg-core) (push) Waiting to run
build container images / core-image-build (ubuntu:22.04, vulkan, true, core, latest-vulkan-ffmpeg-core, --jobs=4 --output-sync=target, linux/amd64, arc-runner-set, false, false, -vulkan-ffmpeg-core) (push) Waiting to run
build container images / gh-runner (nvcr.io/nvidia/l4t-jetpack:r36.4.0, cublas, 12, 0, true, core, latest-nvidia-l4t-arm64-core, --jobs=4 --output-sync=target, linux/arm64, ubuntu-24.04-arm, true, false, -nvidia-l4t-arm64-core) (push) Waiting to run
Security Scan / tests (push) Waiting to run
Tests extras backends / tests-transformers (push) Waiting to run
Tests extras backends / tests-sentencetransformers (push) Waiting to run
Tests extras backends / tests-rerankers (push) Waiting to run
Tests extras backends / tests-diffusers (push) Waiting to run
Tests extras backends / tests-parler-tts (push) Waiting to run
Tests extras backends / tests-openvoice (push) Waiting to run
Tests extras backends / tests-coqui (push) Waiting to run
tests / tests-linux (1.21.x) (push) Waiting to run
tests / tests-aio-container (push) Waiting to run
tests / tests-apple (1.21.x) (push) Waiting to run
chore(docs): extra-Usage and Machine-Tag docs (#4627)
Rename LocalAI-Extra-Usage -> Extra-Usage, add MACHINE_TAG as cli flag option, add docs about extra-usage and machine-tag

Signed-off-by: mintyleaf <mintyleafdev@gmail.com>
2025-01-18 08:58:38 +01:00

213 lines
6.4 KiB
Go

package openai
import (
"bufio"
"bytes"
"encoding/json"
"errors"
"fmt"
"time"
"github.com/mudler/LocalAI/core/backend"
"github.com/mudler/LocalAI/core/config"
"github.com/gofiber/fiber/v2"
"github.com/google/uuid"
"github.com/mudler/LocalAI/core/schema"
"github.com/mudler/LocalAI/pkg/functions"
model "github.com/mudler/LocalAI/pkg/model"
"github.com/mudler/LocalAI/pkg/templates"
"github.com/rs/zerolog/log"
"github.com/valyala/fasthttp"
)
// CompletionEndpoint is the OpenAI Completion API endpoint https://platform.openai.com/docs/api-reference/completions
// @Summary Generate completions for a given prompt and model.
// @Param request body schema.OpenAIRequest true "query params"
// @Success 200 {object} schema.OpenAIResponse "Response"
// @Router /v1/completions [post]
func CompletionEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, evaluator *templates.Evaluator, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
id := uuid.New().String()
created := int(time.Now().Unix())
process := func(s string, req *schema.OpenAIRequest, config *config.BackendConfig, loader *model.ModelLoader, responses chan schema.OpenAIResponse, extraUsage bool) {
ComputeChoices(req, s, config, appConfig, loader, func(s string, c *[]schema.Choice) {}, func(s string, tokenUsage backend.TokenUsage) bool {
usage := schema.OpenAIUsage{
PromptTokens: tokenUsage.Prompt,
CompletionTokens: tokenUsage.Completion,
TotalTokens: tokenUsage.Prompt + tokenUsage.Completion,
}
if extraUsage {
usage.TimingTokenGeneration = tokenUsage.TimingTokenGeneration
usage.TimingPromptProcessing = tokenUsage.TimingPromptProcessing
}
resp := schema.OpenAIResponse{
ID: id,
Created: created,
Model: req.Model, // we have to return what the user sent here, due to OpenAI spec.
Choices: []schema.Choice{
{
Index: 0,
Text: s,
},
},
Object: "text_completion",
Usage: usage,
}
log.Debug().Msgf("Sending goroutine: %s", s)
responses <- resp
return true
})
close(responses)
}
return func(c *fiber.Ctx) error {
// Add Correlation
c.Set("X-Correlation-ID", id)
// Opt-in extra usage flag
extraUsage := c.Get("Extra-Usage", "") != ""
modelFile, input, err := readRequest(c, cl, ml, appConfig, true)
if err != nil {
return fmt.Errorf("failed reading parameters from request:%w", err)
}
log.Debug().Msgf("`input`: %+v", input)
config, input, err := mergeRequestWithConfig(modelFile, input, cl, ml, appConfig.Debug, appConfig.Threads, appConfig.ContextSize, appConfig.F16)
if err != nil {
return fmt.Errorf("failed reading parameters from request:%w", err)
}
if config.ResponseFormatMap != nil {
d := schema.ChatCompletionResponseFormat{}
dat, _ := json.Marshal(config.ResponseFormatMap)
_ = json.Unmarshal(dat, &d)
if d.Type == "json_object" {
input.Grammar = functions.JSONBNF
}
}
config.Grammar = input.Grammar
log.Debug().Msgf("Parameter Config: %+v", config)
if input.Stream {
log.Debug().Msgf("Stream request received")
c.Context().SetContentType("text/event-stream")
//c.Response().Header.SetContentType(fiber.MIMETextHTMLCharsetUTF8)
//c.Set("Content-Type", "text/event-stream")
c.Set("Cache-Control", "no-cache")
c.Set("Connection", "keep-alive")
c.Set("Transfer-Encoding", "chunked")
}
if input.Stream {
if len(config.PromptStrings) > 1 {
return errors.New("cannot handle more than 1 `PromptStrings` when Streaming")
}
predInput := config.PromptStrings[0]
templatedInput, err := evaluator.EvaluateTemplateForPrompt(templates.CompletionPromptTemplate, *config, templates.PromptTemplateData{
Input: predInput,
SystemPrompt: config.SystemPrompt,
})
if err == nil {
predInput = templatedInput
log.Debug().Msgf("Template found, input modified to: %s", predInput)
}
responses := make(chan schema.OpenAIResponse)
go process(predInput, input, config, ml, responses, extraUsage)
c.Context().SetBodyStreamWriter(fasthttp.StreamWriter(func(w *bufio.Writer) {
for ev := range responses {
var buf bytes.Buffer
enc := json.NewEncoder(&buf)
enc.Encode(ev)
log.Debug().Msgf("Sending chunk: %s", buf.String())
fmt.Fprintf(w, "data: %v\n", buf.String())
w.Flush()
}
resp := &schema.OpenAIResponse{
ID: id,
Created: created,
Model: input.Model, // we have to return what the user sent here, due to OpenAI spec.
Choices: []schema.Choice{
{
Index: 0,
FinishReason: "stop",
},
},
Object: "text_completion",
}
respData, _ := json.Marshal(resp)
w.WriteString(fmt.Sprintf("data: %s\n\n", respData))
w.WriteString("data: [DONE]\n\n")
w.Flush()
}))
return nil
}
var result []schema.Choice
totalTokenUsage := backend.TokenUsage{}
for k, i := range config.PromptStrings {
templatedInput, err := evaluator.EvaluateTemplateForPrompt(templates.CompletionPromptTemplate, *config, templates.PromptTemplateData{
SystemPrompt: config.SystemPrompt,
Input: i,
})
if err == nil {
i = templatedInput
log.Debug().Msgf("Template found, input modified to: %s", i)
}
r, tokenUsage, err := ComputeChoices(
input, i, config, appConfig, ml, func(s string, c *[]schema.Choice) {
*c = append(*c, schema.Choice{Text: s, FinishReason: "stop", Index: k})
}, nil)
if err != nil {
return err
}
totalTokenUsage.TimingTokenGeneration += tokenUsage.TimingTokenGeneration
totalTokenUsage.TimingPromptProcessing += tokenUsage.TimingPromptProcessing
result = append(result, r...)
}
usage := schema.OpenAIUsage{
PromptTokens: totalTokenUsage.Prompt,
CompletionTokens: totalTokenUsage.Completion,
TotalTokens: totalTokenUsage.Prompt + totalTokenUsage.Completion,
}
if extraUsage {
usage.TimingTokenGeneration = totalTokenUsage.TimingTokenGeneration
usage.TimingPromptProcessing = totalTokenUsage.TimingPromptProcessing
}
resp := &schema.OpenAIResponse{
ID: id,
Created: created,
Model: input.Model, // we have to return what the user sent here, due to OpenAI spec.
Choices: result,
Object: "text_completion",
Usage: usage,
}
jsonResult, _ := json.Marshal(resp)
log.Debug().Msgf("Response: %s", jsonResult)
// Return the prediction in the response body
return c.JSON(resp)
}
}