Say a transcript's size, name a model by something, and ask before a reload

Five things asked for on the phone, and one trap behind the first of them.

A llama prompt carries `<__media__>` where a picture was, and llama.cpp
pairs each marker with a decoded image when it tokenizes -- so a marker in
words nobody attached a picture to fails the turn, and then fails every
later one, since the conversation is folded out of a transcript that holds
it for good. A model saying the marker back is enough to do it. Every
message with words in it now goes through `without_marker`.

A model's `general.name` is filled in by whatever converted the file, and
`convert_hf_to_gguf.py` fills it from the directory it converted: Prism ML's
Bonsai publishes `general.name = "Hf"`, which is unique and so passed the
label cascade and told a reader nothing. A name is now used only where it
shares a word with the repo or the file it came from; one that does not
drops to the file name.

The model settings dialog and the server card said what a save would cost in
a paragraph under the control, read after the decision if at all. Both ask
instead, in the shape the rest of that screen already uses for Stop and
Delete -- and the dialog's question is asked over the edits, so Cancel comes
back to them.

A field's label was `labelMedium` in the variant colour while every setting
beside it was body text, which on one form read as two ranks of setting.

`GET /sessions/{id}`'s `transcriptFile` now carries the file's size, and the
settings screen draws it beside what this phone has cached.

Checked on the emulator against the sandbox: the labels line up, the row
says "14 kB · 14 kB cached", and both confirmations appear over a loaded
Qwen3-0.6B. `cargo test` 275 passed, clippy and fmt clean, lint clean.
This commit is contained in:
iris-ai committed 2026-09-21 11:59:26 -04:00
1 parent 4b5ed6e398
commit 1aac22bfc9
10 files changed
+290 -51

No files matched your search

+11
View File
@@ -523,6 +523,17 @@ written, and the fold uses that same predicate to decide a reply is settled.
`generate` now fails the turn for both -- a reply that stops early is not a `generate` now fails the turn for both -- a reply that stops early is not a
reply, and the transcript keeps whatever arrived before it. reply, and the transcript keeps whatever arrived before it.
- **`<__media__>` in a llama prompt is a picture, wherever it came from.**
llama.cpp takes an image out of the request and leaves that marker in the
rendered text, then pairs each marker with a decoded image at tokenize time
-- so one in words nobody attached a picture to fails the turn with `number
of media markers in text (1) exceeds number of bitmaps (0)`, which reaches
the phone as "Failed to tokenize prompt". A model saying it back is enough,
and then *every* later message fails too, since the conversation is folded
out of a transcript that now holds it. `Message::new` and
`Message::from_user` take it out of anything that is text (`without_marker`),
which is every message with words in it.
- **A cancel flag is only as prompt as the next place somebody looks.** A - **A cancel flag is only as prompt as the next place somebody looks.** A
llama turn waits on three things that look nowhere at all: a permission llama turn waits on three things that look nowhere at all: a permission
question, a tool call `llama-server` is running (a shell command there runs question, a tool call `llama-server` is running (a shell command there runs
@@ -251,8 +251,18 @@ data class SessionSummary(
val transcriptFile: FileOnMachine?, val transcriptFile: FileOnMachine?,
) )
/** A file somewhere the explorer can be pointed at: which machine, and the path on it. */ /**
data class FileOnMachine(val machine: String, val machineName: String, val path: String) * A file somewhere the explorer can be pointed at: which machine, and the path on it.
*
* [bytes] is how big it is, and null is "the server did not say" rather than zero: a file of no
* bytes and a file nobody could measure are different answers.
*/
data class FileOnMachine(
val machine: String,
val machineName: String,
val path: String,
val bytes: Long? = null,
)
private fun parseSession(session: JSONObject) = private fun parseSession(session: JSONObject) =
SessionSummary( SessionSummary(
@@ -292,6 +302,7 @@ private fun parseSession(session: JSONObject) =
machine = file.getString("machine"), machine = file.getString("machine"),
machineName = file.getString("machineName"), machineName = file.getString("machineName"),
path = file.getString("path"), path = file.getString("path"),
bytes = if (file.has("bytes")) file.getLong("bytes") else null,
) )
}, },
) )
@@ -57,12 +57,10 @@ fun LabelledField(
keyboardActions: KeyboardActions = KeyboardActions.Default, keyboardActions: KeyboardActions = KeyboardActions.Default,
) { ) {
Column(modifier.fillMaxWidth()) { Column(modifier.fillMaxWidth()) {
Text( // Body size in the ordinary text colour, which is what a setting's label is where the
label, // control beside it is a switch or a picker. Smaller and greyer on the ones that are
style = MaterialTheme.typography.labelMedium, // fields reads as two ranks of setting where there is one.
color = MaterialTheme.colorScheme.onSurfaceVariant, Text(label, modifier = Modifier.padding(bottom = 2.dp))
modifier = Modifier.padding(bottom = 2.dp),
)
hint?.let { hint?.let {
Text( Text(
it, it,
@@ -180,12 +180,6 @@ fun ProviderScreen(
if (view.models.isNotEmpty() && view.modelParams.isNotEmpty()) { if (view.models.isNotEmpty() && view.modelParams.isNotEmpty()) {
item("models-heading") { item("models-heading") {
Text("Models", style = MaterialTheme.typography.titleSmall) Text("Models", style = MaterialTheme.typography.titleSmall)
Text(
"How a model is loaded belongs to the machine, not to a session: " +
"one copy of it in memory answers every session using it.",
style = MaterialTheme.typography.bodySmall,
color = MaterialTheme.colorScheme.onSurfaceVariant,
)
Spacer(Modifier.height(8.dp)) Spacer(Modifier.height(8.dp))
} }
} }
@@ -324,6 +318,7 @@ private fun ServerCard(
// half-typed is visibly not saved rather than quietly either way. // half-typed is visibly not saved rather than quietly either way.
val saved = maxLoaded?.toString().orEmpty() val saved = maxLoaded?.toString().orEmpty()
var typed by remember(saved) { mutableStateOf(saved) } var typed by remember(saved) { mutableStateOf(saved) }
var confirming by remember { mutableStateOf(false) }
Card(Modifier.fillMaxWidth()) { Card(Modifier.fillMaxWidth()) {
Column(Modifier.padding(12.dp)) { Column(Modifier.padding(12.dp)) {
Text("Model server", style = MaterialTheme.typography.titleSmall) Text("Model server", style = MaterialTheme.typography.titleSmall)
@@ -354,20 +349,55 @@ private fun ServerCard(
Spacer(Modifier.weight(1f)) Spacer(Modifier.weight(1f))
TextButton( TextButton(
enabled = enabled && typed != saved, enabled = enabled && typed != saved,
onClick = { onMaxLoaded(typed.toIntOrNull()) }, // Saving this while the server is up changes nothing until it comes down
// again, which is asked rather than written underneath -- see [RestartDialog].
onClick = {
if (server.running) confirming = true else onMaxLoaded(typed.toIntOrNull())
},
) { ) {
Text("Save") Text("Save")
} }
} }
if (typed != saved) { }
Text( }
"Read when this server next starts.", if (confirming) {
style = MaterialTheme.typography.bodySmall, RestartDialog(
color = MaterialTheme.colorScheme.onSurfaceVariant, title = "Save for the next start?",
text =
"This server is running, and how many models it keeps loaded was decided when it " +
"started. Saving now changes what it does the next time it starts -- stop it " +
"here to have that be now.",
onConfirm = {
confirming = false
onMaxLoaded(typed.toIntOrNull())
},
onDismiss = { confirming = false },
) )
} }
} }
}
/**
* What a setting read at load time costs, asked before it is written.
*
* One shape for both of the questions on this screen — a model's settings and the server's —
* because they are the same question: this is read when something starts again, and here is what
* starting again costs. A paragraph under the control said the same thing and was read after the
* decision, if at all.
*/
@Composable
private fun RestartDialog(
title: String,
text: String,
onConfirm: () -> Unit,
onDismiss: () -> Unit,
) {
AlertDialog(
onDismissRequest = onDismiss,
title = { Text(title) },
text = { Text(text) },
confirmButton = { TextButton(onClick = onConfirm) { Text("Save") } },
dismissButton = { TextButton(onClick = onDismiss) { Text("Cancel") } },
)
} }
@Composable @Composable
@@ -452,6 +482,10 @@ private fun ModelSettingsDialog(
onSave: (Map<String, String>) -> Unit, onSave: (Map<String, String>) -> Unit,
) { ) {
var params by remember(model.id) { mutableStateOf(model.settings) } var params by remember(model.id) { mutableStateOf(model.settings) }
// Asked over this dialog rather than instead of it, so Cancel comes back to the edits rather
// than throwing them away.
var confirming by remember(model.id) { mutableStateOf(false) }
val loaded = model.status == "loaded" || model.status == "sleeping"
AlertDialog( AlertDialog(
onDismissRequest = onDismiss, onDismissRequest = onDismiss,
// Every control here is a number, so the keyboard is up for most of this dialog's life -- // Every control here is a number, so the keyboard is up for most of this dialog's life --
@@ -463,29 +497,35 @@ private fun ModelSettingsDialog(
title = { Text(model.label) }, title = { Text(model.label) },
text = { text = {
Column(Modifier.verticalScroll(rememberScrollState())) { Column(Modifier.verticalScroll(rememberScrollState())) {
Text(
if (model.status == "loaded" || model.status == "sleeping") {
"This model is loaded. Saving takes it out of memory, and the sessions " +
"using it load it again with these settings on their next message."
} else {
"Read when this model is next loaded."
},
style = MaterialTheme.typography.bodySmall,
color = MaterialTheme.colorScheme.onSurfaceVariant,
)
Spacer(Modifier.height(12.dp))
ProviderParamFields( ProviderParamFields(
specs = specs, specs = specs,
values = params, values = params,
onChange = { params = it }, onChange = { params = it },
// Every one of these is read at load time, and the sentence above already // Every one of these is read at load time, which is what Save stops to say --
// says when that is -- marking each control "on restart" would repeat it six // marking each control "on restart" would repeat it six times over.
// times.
warnAboutRestart = false, warnAboutRestart = false,
) )
} }
}, },
confirmButton = { TextButton(onClick = { onSave(params) }) { Text("Save") } }, confirmButton = {
TextButton(onClick = { if (loaded) confirming = true else onSave(params) }) {
Text("Save")
}
},
dismissButton = { TextButton(onClick = onDismiss) { Text("Cancel") } }, dismissButton = { TextButton(onClick = onDismiss) { Text("Cancel") } },
) )
if (confirming) {
RestartDialog(
title = "Unload ${model.label}?",
text =
"It is in memory now, and these are read when it is loaded. Saving takes it out " +
"of memory; the sessions using it load it again with these settings on their " +
"next message.",
onConfirm = {
confirming = false
onSave(params)
},
onDismiss = { confirming = false },
)
}
} }
@@ -2506,6 +2506,7 @@ fun SessionScreen(
params = params, params = params,
onParamsChanged = { params = it }, onParamsChanged = { params = it },
cachedBytes = cachedBytes, cachedBytes = cachedBytes,
transcriptBytes = summary.transcriptFile?.bytes,
// The purge finishes before the epoch moves, because the relaunched opening effect // The purge finishes before the epoch moves, because the relaunched opening effect
// reads the same directory and would otherwise draw what is about to be deleted. The // reads the same directory and would otherwise draw what is about to be deleted. The
// epoch is what makes the rest a cold open. // epoch is what makes the rest a cold open.
@@ -116,6 +116,11 @@ fun SessionSettingsScreen(
* the Reload row below, which is what would discard it. * the Reload row below, which is what would discard it.
*/ */
cachedBytes: Long?, cachedBytes: Long?,
/**
* How big the record on the server is, or null where it did not say. The other half of the pair
* beside it: what the conversation costs there, against what this phone is holding of it.
*/
transcriptBytes: Long?,
onReload: () -> Unit, onReload: () -> Unit,
/** /**
* Opens the transcript file itself in the explorer. Null from a server that does not say where * Opens the transcript file itself in the explorer. Null from a server that does not say where
@@ -569,24 +574,39 @@ fun SessionSettingsScreen(
modifier = Modifier.fillMaxWidth(), modifier = Modifier.fillMaxWidth(),
) { ) {
Text("Transcript", modifier = Modifier.weight(1f)) Text("Transcript", modifier = Modifier.weight(1f))
// The size is what the button discards, and the unknown state is drawn rather // What the conversation costs on the server, and then what Reload would
// than guessed: a spinner while the directory is being measured, and words when // discard here -- one line, so the two sizes read as a pair. A server that
// there is nothing there, because "nothing cached" and "0 B" read as different // did not measure its file leaves its half out rather than saying zero.
// claims. val onServer = transcriptBytes?.let { humanSize(it) ?: "0 B" }
// The unknown state is drawn rather than guessed: a spinner while the cache
// is being measured, and words when there is nothing in it, because "nothing
// cached" and "0 B" read as different claims.
when { when {
cachedBytes == null -> cachedBytes == null -> {
onServer?.let {
Text(
it,
style = MaterialTheme.typography.bodySmall,
color = MaterialTheme.colorScheme.onSurfaceVariant,
)
Spacer(Modifier.width(8.dp))
}
CircularProgressIndicator( CircularProgressIndicator(
modifier = Modifier.width(16.dp).height(16.dp), modifier = Modifier.width(16.dp).height(16.dp),
strokeWidth = 2.dp, strokeWidth = 2.dp,
) )
else -> }
else -> {
val cached =
humanSize(cachedBytes)?.let { "$it cached" } ?: "nothing cached"
Text( Text(
humanSize(cachedBytes)?.let { "$it cached" } ?: "nothing cached", onServer?.let { "$it · $cached" } ?: cached,
style = MaterialTheme.typography.bodySmall, style = MaterialTheme.typography.bodySmall,
color = MaterialTheme.colorScheme.onSurfaceVariant, color = MaterialTheme.colorScheme.onSurfaceVariant,
) )
} }
} }
}
// Both on a line of their own under what they act on, rather than crowded against // Both on a line of their own under what they act on, rather than crowded against
// the size on the line above: two buttons and a measurement do not fit the width // the size on the line above: two buttons and a measurement do not fit the width
// of a phone, and the one that would lose is the number. // of a phone, and the one that would lose is the number.
+94 -1
View File
@@ -74,7 +74,10 @@ pub struct LocalModel {
pub fn labels(models: &[LocalModel]) -> Vec<String> { pub fn labels(models: &[LocalModel]) -> Vec<String> {
let rungs = |model: &LocalModel| { let rungs = |model: &LocalModel| {
[ [
model.name.clone(), model
.name
.clone()
.filter(|name| names_the_file(name, model)),
Some(model.file.trim_end_matches(".gguf").to_string()), Some(model.file.trim_end_matches(".gguf").to_string()),
Some(model.key.clone()), Some(model.key.clone()),
] ]
@@ -106,6 +109,35 @@ pub fn labels(models: &[LocalModel]) -> Vec<String> {
.collect() .collect()
} }
/// Whether a model's own `general.name` is a name for *this* model, or the
/// converter's working directory wearing the same field.
///
/// The field is filled in by whatever produced the file, and
/// `convert_hf_to_gguf.py` fills it from the directory it converted -- so a
/// model converted out of a folder called `hf` publishes `general.name = "hf"`,
/// which is unique, passes the cascade above, and tells a reader nothing at
/// all. Prism ML's Bonsai is the one here (reported 2026-09-21, drawn as "hf"
/// in the model picker).
///
/// The test is corroboration rather than a list of words to distrust: a real
/// name shares something with where the file came from, both being about the
/// same model, and a directory name picked up in passing does not. One word in
/// common is enough. A name that fails it drops to the next rung, which is the
/// file name the reader downloaded.
fn names_the_file(name: &str, model: &LocalModel) -> bool {
let words = |text: &str| -> Vec<String> {
text.split(|c: char| !c.is_ascii_alphanumeric())
.filter(|word| word.len() > 1)
.map(str::to_ascii_lowercase)
.collect()
};
let from = words(&model.repo);
let from_file = words(&model.file);
words(name)
.iter()
.any(|word| from.contains(word) || from_file.contains(word))
}
/// What a download is doing, or did. /// What a download is doing, or did.
/// ///
/// Flat rather than a tagged enum carrying its message, because the phone /// Flat rather than a tagged enum carrying its message, because the phone
@@ -702,3 +734,64 @@ pub fn urlencode(value: &str) -> String {
}) })
.collect() .collect()
} }
#[cfg(test)]
mod tests {
use super::*;
fn model(repo: &str, file: &str, name: Option<&str>) -> LocalModel {
LocalModel {
key: format!("{repo}/{file}"),
repo: repo.to_string(),
file: file.to_string(),
bytes: 1,
name: name.map(str::to_string),
}
}
/// The cascade, and the one rung that is not simply "is it unique": a name
/// the file it came from says nothing about is the converter's working
/// directory rather than the model's name.
#[test]
fn a_model_is_labelled_by_the_first_thing_that_identifies_it() {
let models = [
// Its own name, which is what a reader recognises.
model(
"Qwen/Qwen3-0.6B-GGUF",
"Qwen3-0.6B-Q8_0.gguf",
Some("Qwen3-0.6B"),
),
// Two quantisations carry one name, so both fall to the file.
model("org/Big-GGUF", "Big-Q4_K_M.gguf", Some("Big")),
model("org/Big-GGUF", "Big-Q8_0.gguf", Some("Big")),
// `convert_hf_to_gguf.py` naming the directory it converted.
model("PrismML/Bonsai-GGUF", "Bonsai-1B-TQ1_0.gguf", Some("hf")),
// Nothing to go on but where it came from.
model("org/Quiet-GGUF", "weights.gguf", None),
];
assert_eq!(
labels(&models),
[
"Qwen3-0.6B",
"Big-Q4_K_M",
"Big-Q8_0",
"Bonsai-1B-TQ1_0",
"weights",
],
);
}
/// Two repos holding a file of the same name: the file rung collides as
/// well, and the key is what always terminates the cascade.
#[test]
fn a_file_name_two_repos_share_falls_through_to_the_key() {
let models = [
model("one/GGUF", "model.gguf", None),
model("two/GGUF", "model.gguf", None),
];
assert_eq!(
labels(&models),
["one/GGUF/model.gguf", "two/GGUF/model.gguf"]
);
}
}
+2 -2
View File
@@ -39,8 +39,8 @@
//! GET /sessions list (id, provider, title, model, status, last activity) //! GET /sessions list (id, provider, title, model, status, last activity)
//! GET /sessions/{id} one session, for refetching after a change. Its //! GET /sessions/{id} one session, for refetching after a change. Its
//! `transcriptFile` is where the record itself is -- //! `transcriptFile` is where the record itself is --
//! {machine, machineName, path} -- so the explorer can be //! {machine, machineName, path, bytes?} -- so the explorer
//! pointed at it //! can be pointed at it and its size said
//! POST /sessions spawn {machine, provider, title?, model?, cwd?, params?} //! POST /sessions spawn {machine, provider, title?, model?, cwd?, params?}
//! POST /sessions/order {sessions} -- the list in the order it is drawn in, //! POST /sessions/order {sessions} -- the list in the order it is drawn in,
//! as the reader dragged it; ids left out keep their //! as the reader dragged it; ids left out keep their
+61 -2
View File
@@ -220,11 +220,46 @@ enum Content {
Parts(Vec<Value>), Parts(Vec<Value>),
} }
/// llama.cpp's placeholder for a picture, as it appears *inside the prompt
/// text* once the server has taken the image out of the request.
///
/// It is an in-band signal in the band that also carries what everybody typed,
/// so it has to be taken back out of anything that is merely words.
/// `mtmd_tokenize` splits the rendered prompt on it and pairs each occurrence
/// with a decoded image, so one nobody sent an image for fails the turn --
/// `number of media markers in text (1) exceeds number of bitmaps (0)`, which
/// reaches the phone as "Failed to tokenize prompt" and then fails every later
/// message too, the words being in the conversation for good. A model that
/// says the marker back is enough to do it.
///
/// llama.cpp's own default, which is what it uses unless started with
/// `--media-marker`; nothing here passes that, and the router writes the preset
/// its children are started from (see [`router`]).
const MEDIA_MARKER: &str = "<__media__>";
/// `text` with any [`MEDIA_MARKER`] removed, for content that is words rather
/// than a picture.
///
/// Removed rather than escaped: there is nothing to escape it *to* -- the
/// marker is matched literally, and the only form llama.cpp will not read as a
/// picture is its absence.
fn without_marker(text: impl Into<String>) -> String {
let text = text.into();
if text.contains(MEDIA_MARKER) {
return text.replace(MEDIA_MARKER, "");
}
text
}
impl Message { impl Message {
/// Words, with anything llama.cpp would read as a picture taken out of
/// them -- see [`without_marker`]. Every message with text goes through
/// here, which is what makes that structural rather than a rule to
/// remember at each call.
fn new(role: &str, content: impl Into<String>) -> Self { fn new(role: &str, content: impl Into<String>) -> Self {
Self { Self {
role: role.to_string(), role: role.to_string(),
content: Content::Text(content.into()), content: Content::Text(without_marker(content)),
tool_calls: Vec::new(), tool_calls: Vec::new(),
tool_call_id: None, tool_call_id: None,
} }
@@ -239,7 +274,7 @@ impl Message {
/// templates, and only the first has ever been tested by everything else /// templates, and only the first has ever been tested by everything else
/// here. /// here.
fn from_user(text: impl Into<String>, images: Vec<Value>) -> Self { fn from_user(text: impl Into<String>, images: Vec<Value>) -> Self {
let text = text.into(); let text = without_marker(text);
if images.is_empty() { if images.is_empty() {
return Self::new("user", text); return Self::new("user", text);
} }
@@ -3252,6 +3287,30 @@ mod tests {
); );
} }
/// llama.cpp's own picture placeholder is not content, and words that
/// contain it are still words: left in, one of them fails the turn it is
/// sent in and every turn after it, because the conversation is folded out
/// of a transcript that now holds it for good.
#[test]
fn a_media_marker_in_words_never_reaches_the_model() {
assert_eq!(
Message::new("user", format!("what is {MEDIA_MARKER} this?")).content,
Content::Text("what is this?".to_string()),
);
// A tool's output is the other text somebody else wrote.
assert_eq!(
Message::result_of("call_1", MEDIA_MARKER).content,
Content::Text(String::new()),
);
// Beside a real picture, which is the part that says there is one.
let image =
json!({"type": "image_url", "image_url": {"url": "data:image/png;base64,UE5H"}});
assert_eq!(
Message::from_user(MEDIA_MARKER, vec![image.clone()]).content,
Content::Parts(vec![image]),
);
}
/// A file that cannot be read is left out rather than failing the turn: /// A file that cannot be read is left out rather than failing the turn:
/// the send path has already said so, and a turn is a worse place to learn /// the send path has already said so, and a turn is a worse place to learn
/// it. /// it.
+9 -3
View File
@@ -342,6 +342,12 @@ pub struct TranscriptFile {
pub machine: String, pub machine: String,
pub machine_name: String, pub machine_name: String,
pub path: String, pub path: String,
/// How big the record is, for a reader deciding whether to open it.
/// Absent where the file could not be measured, which is a session that
/// has not written one yet -- and is deliberately not a zero, since "no
/// transcript" and "we could not look" are different answers.
#[serde(skip_serializing_if = "Option::is_none")]
pub bytes: Option<u64>,
} }
/// Where a session directory keeps its transcript. /// Where a session directory keeps its transcript.
@@ -352,12 +358,12 @@ fn transcript_in(dir: &Path) -> PathBuf {
/// Where `id`'s transcript is, said the way the explorer takes it. /// Where `id`'s transcript is, said the way the explorer takes it.
fn transcript_file(config: &Config, data_dir: &Path, id: &str) -> Option<TranscriptFile> { fn transcript_file(config: &Config, data_dir: &Path, id: &str) -> Option<TranscriptFile> {
let machine = config.machine(crate::config::LOCAL_MACHINE_ID)?; let machine = config.machine(crate::config::LOCAL_MACHINE_ID)?;
let path = transcript_in(&data_dir.join(id));
Some(TranscriptFile { Some(TranscriptFile {
machine: machine.id.clone(), machine: machine.id.clone(),
machine_name: machine.name.clone(), machine_name: machine.name.clone(),
path: transcript_in(&data_dir.join(id)) bytes: std::fs::metadata(&path).ok().map(|file| file.len()),
.to_string_lossy() path: path.to_string_lossy().into_owned(),
.into_owned(),
}) })
} }