o i @sHdZddlZddlZddlmZddlmZddlmZddlm Z ddl Z ddl m Z ddl Z ddlmZmZmZmZdd lmZmZmZmZmZmZmZmZmZmZmZmZm Z m!Z!m"Z"dd l#m$Z$m%Z%m&Z&m'Z'm(Z(m)Z)m*Z*m+Z+m,Z,m-Z-d Z.d Z/d Z0ej1ej2ddgdddZ3Gddde4eZ5 d:de6ej7dej8de9de:efddZ;dedBde4dedede4fd$d%Z?d;ded'e)rTrUfeedbackr`blocked_reasons candidatepartr4r4r5get_parts_by_types:       recCst|d}ddd|DS)z Extract and concatenate all text parts from the response. Args: response: The API response from Gemini. Returns: Combined text from all text parts in the response. rV cSg|]}|jqSr4rV).0rdr4r4r5 z*get_text_from_response..)rejoin)rTr`r4r4r5get_text_from_responses rmFthoughtcsg}t|d}|D])}|jdu|krq |jr%t|jj}tt|}n t|j j IdH}| |q t |dkrT|sOt |}|rKtd|tdtdStj|ddS)Nzimage/*Trz2Gemini did not generate an image. Model response: zGemini did not generate an image. Try rephrasing your prompt or changing the response modality to 'IMAGE+TEXT' to see the model's reasoning.)rp)dim)rernrBbase64 b64decoder@rrrr>r=rElenrmstriprCtorchzeroscat)rTrn image_tensorsr`rd image_datareturned_image model_messager4r4r5get_image_from_responses*     r~cCs"|jsdS|jdvrd}d}d}nJ|jdvrd}d}d}n>|jdvr)d}d}d }n2|jd vr5d }d }d}n&|jd krAd}d}d}n|jdkrMd }d }d}n|jdkrYd}d}d}ndS|jj|}|jjr|jjD]}|jtjkry|||j7}qi|||j7}qi|jjr|||jj7}|dS)N)gemini-2.5-pro-preview-05-06gemini-2.5-prog?g$@g)gemini-2.5-flash-preview-04-17gemini-2.5-flashg333333?g@)r,r-g>@)gemini-3-pro-previewgemini-3.1-pro-previewg(@gemini-3.1-flash-lite-previewg?g?gemini-3-pro-image-previewg^@gemini-3.1-flash-image-previewg?g@gN@g.A) modelVersion usageMetadatapromptTokenCountcandidatesTokensDetailsmodalityrIMAGE tokenCountthoughtsTokenCount)rTinput_tokens_priceoutput_text_tokens_priceoutput_image_tokens_price final_priceir4r4r5calculate_tokens_pricesP          rc@seZdZdZeddZedejdee fddZ edej dee fd d Z e dd e de dedejd Bdej d Bdejd Bdee d Bde dejfddZd S) GeminiNodeaZ Node to generate text responses from a Gemini model. This node allows users to interact with Google's Gemini AI models, providing multimodal inputs (text, images, audio, video, files) to generate coherent text responses. The node works with the latest Gemini models, handling the API communication and response parsing. cCstjddddtjjddddd tjjd gd d d dtjjdddddddtjjddddtjjddddtjjddddt djddddtjjddddd dd!gtj gtj j tj j tj jgdtjtjd gd"d#d$d% S)&Nrz Google Geminiapi node/text/GeminizGenerate text responses with Google's Gemini AI model. You can provide multiple types of inputs (text, images, audio, video) as context for generating more relevant and meaningful responses.promptTzText inputs to the model, used to generate a response. You can include detailed instructions, questions, or context for the model.) multilinedefaulttooltipr$)rrrrrgemini-3-1-progemini-3-1-flash-liter1The Gemini model to use for generating responses.optionsrrseed*r^When seed is fixed to a specific value, the model makes a best effort to provide the same response for repeated requests. Deterministic output isn't guaranteed. Also, changing the model or parameter settings, such as the temperature, can cause variations in the response even when you use the same seed value. By default, a random seed value is used.rrDmaxcontrol_after_generaterr7qOptional image(s) to use as context for the model. To include multiple images, you can use the Batch Images node.optionalraudioz/Optional audio to use as context for the model.videoz/Optional video to use as context for the model.GEMINI_INPUT_FILESfilessOptional file(s) to use as context for the model. Accepts inputs from the Gemini Generate Content Input Files node. system_prompt8Foundational instructions that dictate an AI's behavior.rrrradvancedr&a ( $m := widgets.model; $contains($m, "gemini-2.5-flash") ? { "type": "list_usd", "usd": [0.0003, 0.0025], "format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens"} } : $contains($m, "gemini-2.5-pro") ? { "type": "list_usd", "usd": [0.00125, 0.01], "format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens" } } : ($contains($m, "gemini-3-pro-preview") or $contains($m, "gemini-3-1-pro")) ? { "type": "list_usd", "usd": [0.002, 0.012], "format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens" } } : $contains($m, "gemini-3-1-flash-lite") ? { "type": "list_usd", "usd": [0.00025, 0.0015], "format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens" } } : {"type":"text", "text":"Token-based"} ) r( node_id display_namecategory descriptioninputsoutputshidden is_api_node price_badge)rSchemaStringr ComboIntImageAudioVideoCustomOutputHiddenauth_token_comfy_orgapi_key_comfy_org unique_id PriceBadgePriceBadgeDependsr6r4r4r5 define_schemas   B zGeminiNode.define_schema video_inputr9cCs,t|tjjtjjd}tttj |ddgS)z3Convert video input to Gemini API compatible parts.)container_formatcodecr?rA) r#r VideoContainerMP4 VideoCodecH264rrr video_mp4)r6rbase_64_stringr4r4r5create_video_partsszGeminiNode.create_video_parts audio_inputcCshg}t|djdD]&}tj|d|d|dd}t|ddd}|ttt j |dd q |S) a  Convert audio input to Gemini API compatible parts. Args: audio_input: Audio input from ComfyUI, containing waveform tensor and sample rate. Returns: List of GeminiPart objects containing the encoded audio. waveformr sample_rate)rrmp3 libmp3lame)r codec_namer?rA) rGshaper r unsqueezerrErrr audio_mp3)r6r audio_parts batch_indexaudio_at_index audio_bytesr4r4r5create_audio_partss( zGeminiNode.create_audio_partsNrrr$rr7rrrrc s|dkrd}n |dkrd}n|dkrd}t|dg} |dur*| t||IdH|dur6| |||durB| |||durK| |d} |rYtt|dgdd} t|tt d|d d t t t j | d g| d ttd IdH} t| } t| pdS)Nrrrrrrhr`role/POSTpathmethodrr`)contentssystemInstruction)endpointr@response_modelprice_extractorz#Empty response from Gemini model...)rextendrOrrrrrrGEMINI_BASE_ENDPOINTr r ruserrrrmr NodeOutput) r6rr$rr7rrrrr`gemini_system_promptrT output_textr4r4r5executesF     zGeminiNode.execute)NNNNr)r.r/r0r1 classmethodrr rlistrrrrstrintrrrrr4r4r4r5rsB q!   rc@sXeZdZdZeddZededefddZed d ed e edBde j fd d Z dS)GeminiInputFilesaL Loads and formats input files for use with the Gemini API. This node allows users to include text (.txt) and PDF (.pdf) files as input context for the Gemini model. Files are converted to the appropriate format required by the API and can be chained together to include multiple files in a single request. c Cst}ddt|D}t|ddd}dd|D}tjddd d tjjd ||r.|d nd ddt djddddgt d gdS)z For details about the supported file input types, see: https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/inference cSs>g|]}|r|jds|jdr|jtkr|qS)z.txt.pdf)is_filenameendswithstatst_sizeGEMINI_MAX_INPUT_FILE_SIZErifr4r4r5rjs  z2GeminiInputFiles.define_schema..cSs|jSNr)xr4r4r5 sz0GeminiInputFiles.define_schema..)keycSrgr4r r r4r4r5rj rkrzGemini Input FilesruLoads and prepares input files to include as inputs for Gemini LLM nodes. The files will be read by the Gemini model when generating a response. The contents of the text file count toward the token limit. 🛈 TIP: Can be chained together with other Gemini Input File nodes.filerNzgInput files to include as context for the model. Only accepts text (.txt) and PDF (.pdf) files for now.rrTzAn optional additional file(s) to batch together with the file loaded from this node. Allows chaining of input files so that a single message can include multiple input files.r)rrrrrr) folder_pathsget_input_directoryosscandirsortedrrrr rr)r6 input_dir input_filesr4r4r5rs4  zGeminiInputFiles.define_schema file_pathr9cCsj|drtjntj}t|d }|}Wdn1swYt|d}t t ||ddS)Nrrbzutf-8r?rA) rrapplication_pdf text_plainopenreadrs b64encodedecoderr)r6r mime_typer  file_content base64_strr4r4r5create_file_part's  z!GeminiInputFiles.create_file_partNrrcCs0|durg}t|}||}t|g|S)z-Loads and formats input files for Gemini API.N)rget_annotated_filepathr#rr)r6rrrinput_file_contentr4r4r5r6s   zGeminiInputFiles.executer ) r.r/r0r1rrrrr#rrrrr4r4r4r5rs +(rc@sfeZdZeddZe     ddeded ed ejdBd e e dBd ed edede j fddZ dS) GeminiImagecCstjddddtjjddddd tjjd ttjd d tjjd ddddddtjjddddt djddddtjjdgdddddtjjddd gd!ddd"tjjd#dt dd$dd%gtj tj gtj j tj jtj jgdtjd&d'd( S))NGeminiImageNodez!Nano Banana (Google Gemini Image)api node/image/Geminiz)Edit images synchronously via Google API.rTzText prompt for generationrrrrr$rrrrrrrrr7rrrrr aspect_ratio autoz1:1z2:3z3:2z3:4z4:3z4:5z5:4z9:16z16:9z21:9r,zkDefaults to matching the output image size to that of your input image, or otherwise generates 1:1 squares.)rrrrresponse_modalities IMAGE+TEXTrmChoose 'IMAGE' for image-only output, or 'IMAGE+TEXT' to return both the generated image and a text response.)rrrrrrrzO{"type":"usd","usd":0.039,"format":{"suffix":"/Image (1K)","approximate":true}})r*r)rrrr rr+r3rrrGEMINI_IMAGE_SYS_PROMPTrrrrrrrr4r4r5rBs  ?zGeminiImage.define_schemaNr,r.rrr$rr7rr*r-rr9c st|dddt|dg} |sd}|dkrtnt|d} |dur-| t||IdH|dur6| |d} |rDtt|dgdd} t|td|d d t t t j | d gt |d kr^d gnd d g| d| dttdIdH} tt| IdHt| S)NTrostrip_whitespace min_lengthrhr,) aspectRatior/proxy/vertexai/gemini/rrrrTEXTresponseModalities imageConfigrgenerationConfigrr@rr)r"rrrrOrrrrrr rrrrrrrr~rm) r6rr$rr7rr*r-rr` image_configrrTr4r4r5rs:     zGeminiImage.execute)NNr,r.rr.r/r0rrrrr rrrrrrr4r4r4r5r&@s8 S   r&c@sfeZdZeddZe   ddedededed ed ed ejdBd e e dBd ede j fddZ dS) GeminiImage2cCstjddddtjjddddd tjjd d d gd tjjdddddddtjjdgddddtjjdgdddtjjdddgd dd!tjjd"dd#d$td%jd&dd'd$tjjd(dtdd)dd*g tj tj gtj j tj j tj j gdtd+ S),NGeminiImage2Nodez%Nano Banana Pro (Google Gemini Image)r(r4r4r4r5r?s8 S    r?c@sjeZdZeddZe   ddedededed ed ed ed ejdBd e e dBdede j fddZ dS)GeminiNanoBanana2cCstjddddtjjddddd tjjd d gd tjjd ddddddtjjdgddddtjjdgdddtjjdddgddtjjd d!d"gd tjjd#dd$d%td&jd'dd(d%tjjd)dtdd*dd+g tj tj tjj d,d-d.gtj j tj j tj j gdtd/ S)0NrUz Nano Banana 2r(rArTrBrr)r$rCrDrrrrrErr*r+r,rFrr%rGrKrLr-rr.)rrthinking_levelMINIMALHIGHr7rMrrrrrrr thought_imagezoFirst image from the model's thinking process. Only available with thinking_level HIGH and IMAGE+TEXT modality.)rrrrNrr4r4r5rQs   X zGeminiNanoBanana2.define_schemaNrrr$rr*r%r-rVr7rrr9c  s(t|ddd|dkrd}t|dg} |dur-t|dkr"td| t||IdH| dur6| | t|d } |d krB|| _d} | rPtt | dgdd } t |t d |d dt t tj| dgt|dkrjdgnddg| t|dd| dttdIdH}tt|IdHt|t|ddIdHS)NTror1rCrrhrPrQrRr,rr5rrrrr6) thinkingLevel)r8r9thinkingConfigr:r<)rn)r"rrrCrrOrr4rrrrrr rrrrrrrrr~rm)r6rr$rr*r%r-rVr7rrr`r=rrTr4r4r5rsL        zGeminiNanoBanana2.executerTr>r4r4r4r5rUOs< o     rUc@s(eZdZedeeejfddZdS)GeminiExtensionr9cstttttgSr )rr&r?rUr)selfr4r4r5 get_node_listszGeminiExtension.get_node_listN) r.r/r0rrtyper ComfyNoder^r4r4r4r5r\sr\cstSr )r\r4r4r4r5comfy_entrypointsra)r)F)Jr1rsrenumrriortypingrrwtyping_extensionsrrcomfy_api.latestrrr r comfy_api_nodes.apis.geminir r r rrrrrrrrrrrrcomfy_api_nodes.utilrrrrrrr r!r"r#rrr0rrrOrr+r_r`rrrrOboolrSrermr~floatrrrr&r?rUr\rar4r4r4r5s`     D0    +"+4dO -