# Encoding > UForm for JavaScript Every verb keeps the pooled embeddings by default. Ask for the final-norm states too, one row per token kept, with the `outputs` flags: ```javascript const { embeddings, states, stats } = model.textEncode(texts, arena, { outputs: Outputs.Embeddings | Outputs.States, outputsScalar: Scalar.BF16, }); const tokensOfFirst = states.item(0); // rows from states.offsets[0] up to states.offsets[1] console.log(stats.truncated, stats.tokens); ``` A checkpoint with a token head also lands unit token vectors for late-interaction reranking, `tokenWidth` wide and sharing the states' offsets: ```javascript const { tokenWidth } = model.describe(); // zero without a token head, which answers `unsupported` const { tokenVectors } = model.textEncode(documents, arena, { outputs: Outputs.TokenVectors }); ``` Images, clips and recordings arrive encoded, as JPEG, PNG, WebP, MP4, WebM, WAV, FLAC, MP3 or Opus files: ```javascript import fs from "node:fs/promises"; const images = model.imageEncode([await fs.readFile("panda.jpg")], arena); const clips = model.videoEncode([await fs.readFile("panda.mp4")], arena); const recordings = model.audioEncode([await fs.readFile("panda.wav")], arena); ``` Or decoded, as RGB or RGBA pixels with their shape, and as mono samples with their rate: ```javascript const pixels = model.imageEncodePixels([{ data: rgb, width: 640, height: 480 }], arena); const samples = model.audioEncodeSamples([new Float32Array(16000)], 16000, arena); ``` `import * as uform from 'uform';` `import { Rows } from 'uform';` `import { States } from 'uform';` `import { EncodeVerbs } from 'uform';` ## Outputs The surfaces an encode call keeps, combined with `|`. ```typescript enum Outputs ``` The C verb fills whichever of `embeddings`, `states` and `token_vectors` is non-`NULL`, and these flags decide which it is handed. - `Embeddings` = `1`: One pooled embedding per item. - `States` = `2`: The final-norm states, one row per token kept after truncation. - `TokenVectors` = `4`: The token head's unit vectors, one row per token kept, `tokenWidth` wide; `unsupported` without a token head. ## EncodeConfig Optional per-call settings of every encode verb. ```typescript interface EncodeConfig ``` - `outputs?` `number`: The surfaces to keep; absent keeps [`Outputs.Embeddings`](#Embeddings) alone. - `outputsScalar?` `Scalar`: The scalar every output lands as; absent lands `f32`, which a `Float32Array` reads. - `progress?` `ProgressCallback`: Hears the items encoded so far and may stop the call by answering `false`; only a model's own verbs, which run on the calling thread, take one. ## Pixels One decoded image: `ImageData` as a canvas hands it out, or interleaved RGB or RGBA bytes with their shape. ```typescript interface Pixels ``` - `data` `Uint8Array | Uint8ClampedArray`: Rows end to end, three or four bytes per pixel. - `width` `number`: Pixels per row. - `height` `number`: Rows. ## EncodeStats What an encode call reports beside its outputs. ```typescript interface EncodeStats ``` - `truncated` `number`: Items cut at the model's context. - `tokens` `number`: Tokens the call encoded, after truncation. ## Encoded The surfaces one encode call kept, present as they were asked for. ```typescript interface Encoded ``` - `embeddings?` `Rows`: One pooled embedding per item. - `states?` `States`: The final-norm states of every item. - `tokenVectors?` `States`: The token head's unit vectors of every item, sharing the states' offsets. - `stats` `EncodeStats`: Truncation and token counts. ## Tape A ragged batch packed as the C interface reads it: elements end to end, and `count + 1` offsets into them. ```typescript interface Tape ``` Mirrors `uform_tape_view_t`. Every verb taking an array of entries also takes a packed tape, which the Node addon lends to the call as it is when its offsets are a `BigUint64Array`; an array of entries is first gathered into one tape, the one copy that path makes. A [`Probabilities`](https://www.unum.cloud/docs/uform/javascript/choices.md#Probabilities) answer is itself a tape, so it trains a step's `choiceTargets` as it lands. - `data` `Elements` - `offsets` `Offsets` ## Offsets The `count + 1` offsets of a tape: a `BigUint64Array`, which the Node addon lends as `size_t` without a copy, or a `Uint32Array`, which it widens and the WebAssembly module, whose `size_t` is 32 bits wide, lends as it is. ```typescript type Offsets = BigUint64Array | Uint32Array ``` ## Scalar The scalar kinds a model computes in and its outputs land as, numbered as USearch numbers them. ```typescript enum Scalar ``` - `F64` = `"f64"` - `F32` = `"f32"` - `BF16` = `"bf16"` - `F16` = `"f16"` - `E5M2` = `"e5m2"` - `E4M3` = `"e4m3"` - `E3M2` = `"e3m2"` - `E2M3` = `"e2m3"` - `I8` = `"i8"` - `I4x2` = `"i4x2"` - `I32` = `"i32"` - `U8` = `"u8"` - `U4x2` = `"u4x2"` - `U1x8` = `"u1x8"` ## Rows Rows a call landed in its arena. ```typescript class Rows extends ArenaView ``` Mirrors a `uform_rows_view_t` output. [`data`](https://www.unum.cloud/docs/uform/javascript/choices.md#data) is a `Float32Array` for `f32`, a `Float16Array` for `f16` where the runtime has one, a `Uint16Array` for `bf16`, and raw bytes for the kinds JavaScript spells no element for. ## Rows Made by a verb, never directly. ```typescript new Rows(landed: CompiledRows) ``` ## row Borrows row [`row`](#row) without copying. ```typescript row(row: number): ArrayBufferView ``` Binding-only. ## rows Rows written. ```typescript readonly rows: number ``` ## columns Values in each row, the model's dimensions. ```typescript readonly columns: number ``` ## stride Bytes between consecutive rows. ```typescript readonly stride: number ``` ## scalar The scalar kind the rows landed as. ```typescript readonly scalar: Scalar ``` ## data Every row, laid [`stride`](#stride) bytes apart. ```typescript readonly data: ArrayBufferView ``` ## States The final-norm states or the token vectors a call landed in its arena: rows end to end, item `i` spanning rows `offsets[i]` up to `offsets[i + 1]`. ```typescript class States extends ArenaView ``` Mirrors `uform_rows_tape_view_t`. ## States Made by a verb, never directly. ```typescript new States(landed: CompiledStates) ``` ## item Borrows the rows of item [`item`](https://www.unum.cloud/docs/uform/javascript/choices.md#item) without copying. ```typescript item(item: number): ArrayBufferView ``` Binding-only. ## rows Every item's rows, end to end. ```typescript readonly rows: Rows ``` ## offsets The `count + 1` row offsets, the last of them the total. ```typescript readonly offsets: BigUint64Array ``` ## EncodeVerbs The encode verbs every receiver of a model shares, written once. ```typescript abstract class EncodeVerbs ``` The receiver chooses where a verb runs and how it completes, and the verb stays the same: a [`Model`](https://www.unum.cloud/docs/uform/javascript/models.md#Model) runs it on the calling thread, `model.on(executor)` blocks on an executor, and `model.deferred(executor)` submits it there and answers a [`Deferred`](https://www.unum.cloud/docs/uform/javascript/errors.md#Deferred). Every verb lands its outputs in `arena` and answers views into it. ## textEncode Encodes texts through one forward; every UTF-8 byte is a token. ```typescript textEncode( texts: readonly (string | Uint8Array)[] | Tape, arena: Arena, config: EncodeConfig = {}, ): Answer ``` Calls `uform_text_encode`. - `texts`: Strings, UTF-8 bytes already encoded, or a tape of them. - `arena`: The memory the outputs land in, rewound by the call. - `config`: The surfaces, their scalar, and a progress handler. ## textChoose Scores each item's legal choices through one forward, a softmax over them alone at the item's group's calibrated inverse temperature; a model without the byte decoder reports `unsupported`. ```typescript textChoose( texts: readonly (string | Uint8Array)[] | Tape, choices: readonly (string | Uint8Array)[] | Tape, arena: Arena, config: ChooseConfig = {}, ): Answer ``` Calls `uform_text_choose`. - `texts`: The states, each ending where its choice goes. - `choices`: Each item's legal choices, one byte each, like "yn". - `arena`: The memory the probabilities land in, rewound by the call. - `config`: The question types and a progress handler. ## imageEncode Decodes and encodes JPEG, PNG or WebP files; the browser build reports `unsupported`. ```typescript imageEncode( images: readonly Uint8Array[] | Tape, arena: Arena, config: EncodeConfig = {}, ): Answer ``` Calls `uform_image_encode`. - `images`: One encoded file per item, or a tape of them. - `arena`: The memory the outputs land in, rewound by the call. - `config`: The surfaces, their scalar, and a progress handler. ## imageEncodePixels Encodes decoded images, resized on the executor's side. ```typescript imageEncodePixels( images: readonly Pixels[], arena: Arena, config: EncodeConfig = {}, ): Answer ``` Calls `uform_image_encode_pixels`. - `images`: `ImageData`, or RGB or RGBA bytes with their width and height. - `arena`: The memory the outputs land in, rewound by the call. - `config`: The surfaces, their scalar, and a progress handler. ## videoEncode Decodes and encodes MP4 or WebM clips; the browser build reports `unsupported`. ```typescript videoEncode( videos: readonly Uint8Array[] | Tape, arena: Arena, config: EncodeConfig = {}, ): Answer ``` Calls `uform_video_encode`. - `videos`: One encoded file per item, or a tape of them. - `arena`: The memory the outputs land in, rewound by the call. - `config`: The surfaces, their scalar, and a progress handler. ## audioEncode Decodes and encodes WAV, FLAC, MP3 or Opus files; the browser build reports `unsupported`. ```typescript audioEncode( recordings: readonly Uint8Array[] | Tape, arena: Arena, config: EncodeConfig = {}, ): Answer ``` Calls `uform_audio_encode`. - `recordings`: One encoded file per item, or a tape of them. - `arena`: The memory the outputs land in, rewound by the call. - `config`: The surfaces, their scalar, and a progress handler. ## audioEncodeSamples Encodes decoded mono recordings, resampled to the model's rate where `sampleRate` differs. ```typescript audioEncodeSamples( recordings: readonly Float32Array[] | Tape, sampleRate: number, arena: Arena, config: EncodeConfig = {}, ): Answer ``` Calls `uform_audio_encode_samples`. - `recordings`: One run of samples per item, or a tape of them. - `sampleRate`: Samples per second shared by every recording. - `arena`: The memory the outputs land in, rewound by the call. - `config`: The surfaces, their scalar, and a progress handler. ## executor The executor the verbs run on; absent runs them on the calling thread. ```typescript readonly executor: Executor | undefined ```