chiyi/image2api
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

加回 nano-banana-2 与 nano-banana-pro 并存

Browse Source
This commit is contained in:
chiyi
2026-07-10 12:02:14 +08:00
parent 2afb4e7857
commit 8d792e66e0
3 changed files with 49 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/internal/http/handler/user_generation.go
+20
View File
@@ -452,6 +452,17 @@ func (h *UserGenerationHandler) catalogEntries(c *gin.Context) ([]gin.H, error)
"max_reference_images": 4,
"description": "Adobe Flux Kontext Max",
},
{
"id": "nano-banana-2",
"provider": "runway",
"type": "image",
"ratios": []string{"1:1", "1:4", "1:8", "2:3", "3:2", "3:4", "4:1", "4:3", "4:5", "5:4", "8:1", "9:16", "16:9", "21:9"},
"resolutions": []string{"1K", "2K", "4K"},
"image_to_image": true,
"max_reference_images": 6,
"reference_mode": "asset",
"description": "Runway Nano Banana 2 (图/参考图)",
},
{
"id": "nano-banana-pro",
"provider": "runway",
@@ -613,6 +624,15 @@ func (h *UserGenerationHandler) publicModels() ([]gin.H, error) {
"description": "Adobe Flux Kontext Max",
"stub": false,
},
{
"id": "nano-banana-2",
"provider": "runway",
"kind": "image",
"ratios": []string{"1:1", "1:4", "1:8", "2:3", "3:2", "3:4", "4:1", "4:3", "4:5", "5:4", "8:1", "9:16", "16:9", "21:9"},
"resolutions": []string{"1K", "2K", "4K"},
"description": "Runway Nano Banana 2",
"stub": false,
},
{
"id": "nano-banana-pro",
"provider": "runway",
backend/internal/provider/runway/image.go
+24 -13
View File
@@ -13,14 +13,15 @@ import (
"github.com/google/uuid"
)
// GenerateImage runs the Runway "Nano Banana Pro" (workflow_gemini_image /
// gemini-3-pro-image-preview) text/image-to-image pipeline: upload each
// reference image (DATASET + DATASET_PREVIEW → dataset) to obtain its
// {assetId, url}, create a gemini image task and poll it to completion, then
// download the rendered PNG. teamID is the workspace id; if empty it's derived
// from the token. imageSize is the "1K"/"2K"/"4K" tier (Pro honors 2K/4K).
// refs may be empty (pure text-to-image).
func (c *Client) GenerateImage(ctx context.Context, token, teamID, prompt, imageSize string, refs [][]byte) ([]byte, map[string]any, error) {
// GenerateImage runs a Runway gemini image text/image-to-image pipeline:
// upload each reference image (DATASET + DATASET_PREVIEW → dataset) to obtain
// its {assetId, url}, create a gemini image task and poll it to completion,
// then download the rendered PNG. modelID selects the variant: "nano-banana-2"
// → gemini_3_1_flash_image / gemini-3.1-flash-image-preview (with aspect_ratio),
// anything else → workflow_gemini_image / gemini-3-pro-image-preview. imageSize
// is the "1K"/"2K"/"4K" tier. teamID is the workspace id; if empty it's derived
// from the token. refs may be empty (pure text-to-image).
func (c *Client) GenerateImage(ctx context.Context, token, teamID, modelID, prompt, aspectRatio, imageSize string, refs [][]byte) ([]byte, map[string]any, error) {
token = strings.TrimSpace(strings.TrimPrefix(token, "Bearer "))
if token == "" {
return nil, nil, ErrAuth
@@ -63,7 +64,7 @@ func (c *Client) GenerateImage(ctx context.Context, token, teamID, prompt, image
})
}
taskID, err := c.createImageTask(ctx, submitClient, token, teamID, prompt, imageSize, refImages)
taskID, err := c.createImageTask(ctx, submitClient, token, teamID, modelID, prompt, aspectRatio, imageSize, refImages)
if err != nil {
return nil, nil, err
}
@@ -108,9 +109,10 @@ func (c *Client) uploadReference(ctx context.Context, client tlsclient.HttpClien
return assetID, refURL, nil
}
// createImageTask creates a workflow_gemini_image (Nano Banana Pro) task and
// returns its id.
func (c *Client) createImageTask(ctx context.Context, client tlsclient.HttpClient, token, teamID, prompt, imageSize string, refImages []map[string]any) (string, error) {
// createImageTask creates a gemini image task (workflow_gemini_image for Pro,
// gemini_3_1_flash_image for Nano Banana 2) and returns its id.
func (c *Client) createImageTask(ctx context.Context, client tlsclient.HttpClient, token, teamID, modelID, prompt, aspectRatio, imageSize string, refImages []map[string]any) (string, error) {
taskType := "workflow_gemini_image"
opts := map[string]any{
"name": "Nano Banana Pro - " + prompt,
"text_prompt": prompt,
@@ -120,11 +122,20 @@ func (c *Client) createImageTask(ctx context.Context, client tlsclient.HttpClien
"exploreMode": false,
"creationSource": "tool-mode",
}
if modelID == "nano-banana-2" {
taskType = "gemini_3_1_flash_image"
opts["name"] = "Nano Banana 2 - " + prompt
opts["model"] = "gemini-3.1-flash-image-preview"
if strings.TrimSpace(aspectRatio) == "" {
aspectRatio = "16:9"
}
opts["aspect_ratio"] = aspectRatio
}
if len(refImages) > 0 {
opts["reference_images"] = refImages
}
res, err := c.submitTask(ctx, client, token, teamID, map[string]any{
"taskType": "workflow_gemini_image",
"taskType": taskType,
"options": opts,
"asTeamId": jsonNumberOrString(teamID),
"sessionId": uuid.NewString(),
backend/internal/service/v1.go
+5 -5
View File
@@ -555,7 +555,7 @@ func (s *V1Service) prepareImageExecution(ctx context.Context, principal *APIPri
}
imageBytes = b
case "runway":
b, execErr := s.generateRunwayImage(genCtx, eventID, modelItem, in, resolution)
b, execErr := s.generateRunwayImage(genCtx, eventID, modelItem, in, aspectRatio, resolution)
if execErr != nil {
_ = s.refundIfNeeded(ctx, principal, eventID, price)
_ = s.events.UpdateStatus(ctx, eventID, "failed", execErr.Error(), 0)
@@ -2137,14 +2137,14 @@ func (s *V1Service) generateGrokVideo(ctx context.Context, eventID string, model
return nil, "", lastErr
}
// generateRunwayImage runs the Runway "Nano Banana Pro" (workflow_gemini_image /
// gemini-3-pro-image-preview) image pipeline across the runway pool. Unlike the
// generateRunwayImage runs the Runway gemini image pipeline (Nano Banana Pro or
// Nano Banana 2, selected by the model id) across the runway pool. Unlike the
// video path it does NOT pre-deduct credits: it simply round-robins the pool and
// generates. Per ops decision an out-of-credits account is treated like a dead
// 401 — marked dead (status=disabled) and skipped — because Runway credits don't
// refill daily, so a "quota" mark (which the maintenance loop would revive) is
// wrong. Reference images (up to the model's max) are uploaded per attempt.
func (s *V1Service) generateRunwayImage(ctx context.Context, eventID string, modelItem *model.ModelConfig, in V1ImageRequest, resolution string) ([]byte, error) {
func (s *V1Service) generateRunwayImage(ctx context.Context, eventID string, modelItem *model.ModelConfig, in V1ImageRequest, aspectRatio, resolution string) ([]byte, error) {
if s.runway == nil {
return nil, errors.New("runway client not configured")
}
@@ -2203,7 +2203,7 @@ func (s *V1Service) generateRunwayImage(ctx context.Context, eventID string, mod
if token.Meta != nil {
teamID = strings.TrimSpace(stringValue(token.Meta["team_id"]))
}
d, _, genErr := s.runway.GenerateImage(ctx, token.Value, teamID, in.Prompt, imageSize, refs)
d, _, genErr := s.runway.GenerateImage(ctx, token.Value, teamID, modelItem.ID, in.Prompt, aspectRatio, imageSize, refs)
if genErr == nil {
_, _ = s.tokens.Update(ctx, "runway", token.ID, map[string]any{
"last_used_at": time.Now(),