Spaces:
Running
Running
Sync from felladrin/MiniSearch@b72789d
Browse files- agents.md +1 -1
- changelog.md +5 -0
- client/modules/textGenerationWithWllama.test.ts +24 -1
- client/modules/textGenerationWithWllama.ts +26 -3
- client/modules/wllama.test.ts +6 -10
- client/modules/wllama.ts +12 -9
- docs/ai-integration.md +1 -1
- docs/configuration.md +1 -1
agents.md
CHANGED
|
@@ -190,7 +190,7 @@ Need to:
|
|
| 190 |
|
| 191 |
1. Check the wasm build can run it: the architecture must exist in `node_modules/@wllama/wllama/esm/wasm/wllama.wasm` (grep the binary for the arch name), and the quant must stay under the ceilings in `docs/ai-integration.md` (shard every file under 2000 MiB; total weights under ~3.7 GB)
|
| 192 |
2. Shard the GGUF (`llama-gguf-split --split --size`, or an equivalent) and publish to a `Felladrin/gguf-sharded-<QUANT>-<Name>` repo
|
| 193 |
-
3. Add to `client/modules/wllama.ts`; thinking-by-default templates need `
|
| 194 |
4. Run `node scripts/wllama-catalog-check.cjs` and `npx vitest run client/modules/wllama.test.ts`
|
| 195 |
5. Update `docs/ai-integration.md`
|
| 196 |
6. Update `docs/configuration.md` defaults
|
|
|
|
| 190 |
|
| 191 |
1. Check the wasm build can run it: the architecture must exist in `node_modules/@wllama/wllama/esm/wasm/wllama.wasm` (grep the binary for the arch name), and the quant must stay under the ceilings in `docs/ai-integration.md` (shard every file under 2000 MiB; total weights under ~3.7 GB)
|
| 192 |
2. Shard the GGUF (`llama-gguf-split --split --size`, or an equivalent) and publish to a `Felladrin/gguf-sharded-<QUANT>-<Name>` repo
|
| 193 |
+
3. Add to `client/modules/wllama.ts`; thinking-by-default templates need nothing extra, their `reasoning_content` channel is streamed into the thinking UI
|
| 194 |
4. Run `node scripts/wllama-catalog-check.cjs` and `npx vitest run client/modules/wllama.test.ts`
|
| 195 |
5. Update `docs/ai-integration.md`
|
| 196 |
6. Update `docs/configuration.md` defaults
|
changelog.md
CHANGED
|
@@ -3,6 +3,11 @@
|
|
| 3 |
Notable changes, grouped by date, most recent first.
|
| 4 |
|
| 5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
## 2026-11-02
|
| 7 |
|
| 8 |
- Five new browser models joined the Wllama catalog: Spark X2.6 1.7B, Granite 4.2 3B, Qwen 3.5 4B, Gemma 4 E2B and Spark X2.6 4B, each tested end to end against the shipped wasm build with the search-answer prompt. The Gemma 4 and Qwen 3.5 architectures were previously absent from the catalog, and the new entries run on the wasm engine's supported architectures (`gemma4`, `qwen35`, `spark2_5`, `granite`). Models whose chat template thinks by default ship with `chatTemplateKwargs: { enable_thinking: false }`, a new per-model field passed to wllama as `chat_template_kwargs`, because MiniSearch renders only the content channel and a thinking template would otherwise spend the whole token budget on reasoning and answer with nothing. `docs/ai-integration.md` now records the two measured ceilings a candidate must fit (shard files under 2000 MiB, since Chromium's OPFS silently truncates beyond it; total weights under ~3.7 GB, since the wasm linear memory caps at 4 GiB), and `npm run wllama-catalog-check` HEADs every shard of every catalog model on HuggingFace to verify the claimed sizes.
|
|
|
|
| 3 |
Notable changes, grouped by date, most recent first.
|
| 4 |
|
| 5 |
|
| 6 |
+
## 2026-11-03
|
| 7 |
+
|
| 8 |
+
- The five thinking models added on 2026-11-02 (Spark X2.6 1.7B and 4B, Granite 4.2 3B, Qwen 3.5 4B, Gemma 4 E2B) no longer ship with thinking switched off. Their chain of thought arrives on wllama's `reasoning_content` channel and is now wrapped in the configured reasoning markers, so the existing thinking UI shows it while the model reasons and the answer streams in after. The model warmup counts a reasoning token as a working model, so thinking models no longer log a failed warmup.
|
| 9 |
+
|
| 10 |
+
|
| 11 |
## 2026-11-02
|
| 12 |
|
| 13 |
- Five new browser models joined the Wllama catalog: Spark X2.6 1.7B, Granite 4.2 3B, Qwen 3.5 4B, Gemma 4 E2B and Spark X2.6 4B, each tested end to end against the shipped wasm build with the search-answer prompt. The Gemma 4 and Qwen 3.5 architectures were previously absent from the catalog, and the new entries run on the wasm engine's supported architectures (`gemma4`, `qwen35`, `spark2_5`, `granite`). Models whose chat template thinks by default ship with `chatTemplateKwargs: { enable_thinking: false }`, a new per-model field passed to wllama as `chat_template_kwargs`, because MiniSearch renders only the content channel and a thinking template would otherwise spend the whole token budget on reasoning and answer with nothing. `docs/ai-integration.md` now records the two measured ceilings a candidate must fit (shard files under 2000 MiB, since Chromium's OPFS silently truncates beyond it; total weights under ~3.7 GB, since the wasm linear memory caps at 4 GiB), and `npm run wllama-catalog-check` HEADs every shard of every catalog model on HuggingFace to verify the claimed sizes.
|
client/modules/textGenerationWithWllama.test.ts
CHANGED
|
@@ -3,7 +3,10 @@ import {
|
|
| 3 |
updateModelLoadingProgress,
|
| 4 |
updateModelSizeInMegabytes,
|
| 5 |
} from "./pubSub";
|
| 6 |
-
import {
|
|
|
|
|
|
|
|
|
|
| 7 |
import { initializeWllama } from "./wllama";
|
| 8 |
|
| 9 |
vi.mock("./logEntries", () => ({ addLogEntry: vi.fn() }));
|
|
@@ -14,6 +17,8 @@ vi.mock("./pubSub", () => ({
|
|
| 14 |
enableAiResponse: true,
|
| 15 |
wllamaModelId: "test-model",
|
| 16 |
cpuThreads: 1,
|
|
|
|
|
|
|
| 17 |
}),
|
| 18 |
getTextGenerationState: vi.fn().mockReturnValue("generating"),
|
| 19 |
updateModelLoadingProgress: vi.fn(),
|
|
@@ -106,3 +111,21 @@ describe("generateTextWithWllama", () => {
|
|
| 106 |
expect(updateModelSizeInMegabytes).not.toHaveBeenCalledWith(0);
|
| 107 |
});
|
| 108 |
});
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
updateModelLoadingProgress,
|
| 4 |
updateModelSizeInMegabytes,
|
| 5 |
} from "./pubSub";
|
| 6 |
+
import {
|
| 7 |
+
composeStreamedText,
|
| 8 |
+
generateTextWithWllama,
|
| 9 |
+
} from "./textGenerationWithWllama";
|
| 10 |
import { initializeWllama } from "./wllama";
|
| 11 |
|
| 12 |
vi.mock("./logEntries", () => ({ addLogEntry: vi.fn() }));
|
|
|
|
| 17 |
enableAiResponse: true,
|
| 18 |
wllamaModelId: "test-model",
|
| 19 |
cpuThreads: 1,
|
| 20 |
+
reasoningStartMarker: "<think>",
|
| 21 |
+
reasoningEndMarker: "</think>",
|
| 22 |
}),
|
| 23 |
getTextGenerationState: vi.fn().mockReturnValue("generating"),
|
| 24 |
updateModelLoadingProgress: vi.fn(),
|
|
|
|
| 111 |
expect(updateModelSizeInMegabytes).not.toHaveBeenCalledWith(0);
|
| 112 |
});
|
| 113 |
});
|
| 114 |
+
|
| 115 |
+
describe("composeStreamedText", () => {
|
| 116 |
+
it("returns the answer unchanged when the model does not think", () => {
|
| 117 |
+
expect(composeStreamedText("", "The answer.")).toBe("The answer.");
|
| 118 |
+
});
|
| 119 |
+
|
| 120 |
+
it("keeps the reasoning open while the answer has not started", () => {
|
| 121 |
+
expect(composeStreamedText("hmm, let me see", "")).toBe(
|
| 122 |
+
"<think>hmm, let me see",
|
| 123 |
+
);
|
| 124 |
+
});
|
| 125 |
+
|
| 126 |
+
it("closes the reasoning block when the answer starts", () => {
|
| 127 |
+
expect(composeStreamedText("done thinking", "The answer.")).toBe(
|
| 128 |
+
"<think>done thinking</think>The answer.",
|
| 129 |
+
);
|
| 130 |
+
});
|
| 131 |
+
});
|
client/modules/textGenerationWithWllama.ts
CHANGED
|
@@ -113,6 +113,8 @@ async function generateWithWllama({
|
|
| 113 |
? await messagesOrBuilder()
|
| 114 |
: messagesOrBuilder;
|
| 115 |
|
|
|
|
|
|
|
| 116 |
let streamedMessage = "";
|
| 117 |
|
| 118 |
const stream = await wllama.createChatCompletion({
|
|
@@ -135,16 +137,23 @@ async function generateWithWllama({
|
|
| 135 |
}
|
| 136 |
|
| 137 |
const delta = chunk.choices[0]?.delta?.content ?? "";
|
| 138 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 139 |
|
| 140 |
if (shouldCheckCanRespond && getTextGenerationState() !== "generating") {
|
| 141 |
updateTextGenerationState("generating");
|
| 142 |
}
|
| 143 |
|
| 144 |
-
|
|
|
|
|
|
|
| 145 |
streamedMessage = handleWllamaCompletion(
|
| 146 |
model,
|
| 147 |
-
|
| 148 |
() => abortController.abort(),
|
| 149 |
onUpdate,
|
| 150 |
);
|
|
@@ -199,6 +208,20 @@ async function initializeWllamaInstance(progressCallback?: ProgressCallback) {
|
|
| 199 |
return { wllama, model };
|
| 200 |
}
|
| 201 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
function handleWllamaCompletion(
|
| 203 |
model: WllamaModel,
|
| 204 |
currentText: string,
|
|
|
|
| 113 |
? await messagesOrBuilder()
|
| 114 |
: messagesOrBuilder;
|
| 115 |
|
| 116 |
+
let reasoningText = "";
|
| 117 |
+
let answerText = "";
|
| 118 |
let streamedMessage = "";
|
| 119 |
|
| 120 |
const stream = await wllama.createChatCompletion({
|
|
|
|
| 137 |
}
|
| 138 |
|
| 139 |
const delta = chunk.choices[0]?.delta?.content ?? "";
|
| 140 |
+
// wllama splits a thinking model's chain of thought into its own
|
| 141 |
+
// channel; the type does not declare it yet.
|
| 142 |
+
const reasoningDelta =
|
| 143 |
+
(chunk.choices[0]?.delta as { reasoning_content?: string } | undefined)
|
| 144 |
+
?.reasoning_content ?? "";
|
| 145 |
+
if (!delta && !reasoningDelta) continue;
|
| 146 |
|
| 147 |
if (shouldCheckCanRespond && getTextGenerationState() !== "generating") {
|
| 148 |
updateTextGenerationState("generating");
|
| 149 |
}
|
| 150 |
|
| 151 |
+
if (reasoningDelta) reasoningText += reasoningDelta;
|
| 152 |
+
if (delta) answerText += delta;
|
| 153 |
+
|
| 154 |
streamedMessage = handleWllamaCompletion(
|
| 155 |
model,
|
| 156 |
+
composeStreamedText(reasoningText, answerText),
|
| 157 |
() => abortController.abort(),
|
| 158 |
onUpdate,
|
| 159 |
);
|
|
|
|
| 208 |
return { wllama, model };
|
| 209 |
}
|
| 210 |
|
| 211 |
+
/**
|
| 212 |
+
* Joins the reasoning and answer channels into one stream so the existing
|
| 213 |
+
* reasoning-marker parsing in the UI shows the chain of thought while the
|
| 214 |
+
* model thinks. The end marker only appears once the answer has started,
|
| 215 |
+
* which keeps the UI in its "still thinking" state until then.
|
| 216 |
+
*/
|
| 217 |
+
export function composeStreamedText(reasoning: string, answer: string): string {
|
| 218 |
+
if (!reasoning) return answer;
|
| 219 |
+
const { reasoningStartMarker, reasoningEndMarker } = getSettings();
|
| 220 |
+
return `${reasoningStartMarker}${reasoning}${
|
| 221 |
+
answer ? `${reasoningEndMarker}${answer}` : ""
|
| 222 |
+
}`;
|
| 223 |
+
}
|
| 224 |
+
|
| 225 |
function handleWllamaCompletion(
|
| 226 |
model: WllamaModel,
|
| 227 |
currentText: string,
|
client/modules/wllama.test.ts
CHANGED
|
@@ -36,17 +36,13 @@ describe("wllamaModels catalog", () => {
|
|
| 36 |
}
|
| 37 |
});
|
| 38 |
|
| 39 |
-
it("
|
| 40 |
-
//
|
| 41 |
-
//
|
| 42 |
-
//
|
| 43 |
-
//
|
| 44 |
for (const [id, model] of models) {
|
| 45 |
-
|
| 46 |
-
expect(model.chatTemplateKwargs, id).toEqual({
|
| 47 |
-
enable_thinking: false,
|
| 48 |
-
});
|
| 49 |
-
}
|
| 50 |
}
|
| 51 |
});
|
| 52 |
|
|
|
|
| 36 |
}
|
| 37 |
});
|
| 38 |
|
| 39 |
+
it("keeps thinking enabled on every model", () => {
|
| 40 |
+
// The reasoning channel is streamed into the thinking UI, so a model
|
| 41 |
+
// must not hand the template a kwarg that switches thinking off: the
|
| 42 |
+
// answer would arrive with no visible chain of thought, which reads as
|
| 43 |
+
// a dumber model rather than a feature that was turned off.
|
| 44 |
for (const [id, model] of models) {
|
| 45 |
+
expect(model.chatTemplateKwargs?.enable_thinking, id).not.toBe(false);
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
}
|
| 47 |
});
|
| 48 |
|
client/modules/wllama.ts
CHANGED
|
@@ -70,7 +70,15 @@ export async function initializeWllama(
|
|
| 70 |
: {}),
|
| 71 |
});
|
| 72 |
|
| 73 |
-
const
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
|
| 75 |
addLogEntry(
|
| 76 |
`Wllama warmup ${hasWarmupSucceeded ? "succeeded" : "failed"}.`,
|
|
@@ -128,9 +136,9 @@ export interface WllamaModel {
|
|
| 128 |
readonly flash_attn?: boolean;
|
| 129 |
/**
|
| 130 |
* Extra variables handed to the model's chat template on every completion
|
| 131 |
-
* (wllama's `chat_template_kwargs`).
|
| 132 |
-
*
|
| 133 |
-
*
|
| 134 |
*/
|
| 135 |
readonly chatTemplateKwargs?: Record<string, unknown>;
|
| 136 |
getSampling: () => Omit<SamplingParams, "logit_bias">;
|
|
@@ -337,7 +345,6 @@ export const wllamaModels: Readonly<Record<string, WllamaModel>> = {
|
|
| 337 |
hfRepoId: "Felladrin/gguf-sharded-Q4_K_M-Spark-X2.6-1.7B",
|
| 338 |
hfFilePath: "Spark-X2.6-1.7B-Q4_K_M-00001-of-00004.gguf",
|
| 339 |
fileSizeInMegabytes: 1769,
|
| 340 |
-
chatTemplateKwargs: { enable_thinking: false },
|
| 341 |
},
|
| 342 |
"smollm3-3b": {
|
| 343 |
...createDefaultModelConfig(),
|
|
@@ -373,7 +380,6 @@ export const wllamaModels: Readonly<Record<string, WllamaModel>> = {
|
|
| 373 |
hfRepoId: "Felladrin/gguf-sharded-Q4_K_S-granite-4.2-3b",
|
| 374 |
hfFilePath: "granite-4.2-3b-Q4_K_S-00001-of-00005.gguf",
|
| 375 |
fileSizeInMegabytes: 2143,
|
| 376 |
-
chatTemplateKwargs: { enable_thinking: false },
|
| 377 |
},
|
| 378 |
"phi-3.5-mini-3.8b": {
|
| 379 |
...createDefaultModelConfig(),
|
|
@@ -402,7 +408,6 @@ export const wllamaModels: Readonly<Record<string, WllamaModel>> = {
|
|
| 402 |
hfRepoId: "Felladrin/gguf-sharded-UD-Q3_K_XL-Qwen3.5-4B",
|
| 403 |
hfFilePath: "Qwen3.5-4B-UD-Q3_K_XL-00001-of-00005.gguf",
|
| 404 |
fileSizeInMegabytes: 2436,
|
| 405 |
-
chatTemplateKwargs: { enable_thinking: false },
|
| 406 |
},
|
| 407 |
"gemma-4-e2b": {
|
| 408 |
...createDefaultModelConfig(),
|
|
@@ -410,7 +415,6 @@ export const wllamaModels: Readonly<Record<string, WllamaModel>> = {
|
|
| 410 |
hfRepoId: "Felladrin/gguf-sharded-Q3_K_S-gemma-4-e2b-it",
|
| 411 |
hfFilePath: "gemma-4-e2b-it-Q3_K_S-00001-of-00005.gguf",
|
| 412 |
fileSizeInMegabytes: 2446,
|
| 413 |
-
chatTemplateKwargs: { enable_thinking: false },
|
| 414 |
},
|
| 415 |
"phi-4-mini-reasoning": {
|
| 416 |
...createDefaultModelConfig(),
|
|
@@ -425,6 +429,5 @@ export const wllamaModels: Readonly<Record<string, WllamaModel>> = {
|
|
| 425 |
hfRepoId: "Felladrin/gguf-sharded-Q4_K_M-Spark-X2.6-4B",
|
| 426 |
hfFilePath: "Spark-X2.6-4B-Q4_K_M-00001-of-00007.gguf",
|
| 427 |
fileSizeInMegabytes: 3543,
|
| 428 |
-
chatTemplateKwargs: { enable_thinking: false },
|
| 429 |
},
|
| 430 |
};
|
|
|
|
| 70 |
: {}),
|
| 71 |
});
|
| 72 |
|
| 73 |
+
const warmupMessage = warmupResponse.choices[0]?.message as
|
| 74 |
+
| { content?: string; reasoning_content?: string }
|
| 75 |
+
| undefined;
|
| 76 |
+
|
| 77 |
+
// A thinking model spends its single warmup token on the reasoning
|
| 78 |
+
// channel, so either channel counts as a working model.
|
| 79 |
+
const hasWarmupSucceeded = !!(
|
| 80 |
+
warmupMessage?.content || warmupMessage?.reasoning_content
|
| 81 |
+
);
|
| 82 |
|
| 83 |
addLogEntry(
|
| 84 |
`Wllama warmup ${hasWarmupSucceeded ? "succeeded" : "failed"}.`,
|
|
|
|
| 136 |
readonly flash_attn?: boolean;
|
| 137 |
/**
|
| 138 |
* Extra variables handed to the model's chat template on every completion
|
| 139 |
+
* (wllama's `chat_template_kwargs`). Thinking stays on: the reasoning
|
| 140 |
+
* channel is streamed into the thinking UI, so nothing here should turn
|
| 141 |
+
* it off.
|
| 142 |
*/
|
| 143 |
readonly chatTemplateKwargs?: Record<string, unknown>;
|
| 144 |
getSampling: () => Omit<SamplingParams, "logit_bias">;
|
|
|
|
| 345 |
hfRepoId: "Felladrin/gguf-sharded-Q4_K_M-Spark-X2.6-1.7B",
|
| 346 |
hfFilePath: "Spark-X2.6-1.7B-Q4_K_M-00001-of-00004.gguf",
|
| 347 |
fileSizeInMegabytes: 1769,
|
|
|
|
| 348 |
},
|
| 349 |
"smollm3-3b": {
|
| 350 |
...createDefaultModelConfig(),
|
|
|
|
| 380 |
hfRepoId: "Felladrin/gguf-sharded-Q4_K_S-granite-4.2-3b",
|
| 381 |
hfFilePath: "granite-4.2-3b-Q4_K_S-00001-of-00005.gguf",
|
| 382 |
fileSizeInMegabytes: 2143,
|
|
|
|
| 383 |
},
|
| 384 |
"phi-3.5-mini-3.8b": {
|
| 385 |
...createDefaultModelConfig(),
|
|
|
|
| 408 |
hfRepoId: "Felladrin/gguf-sharded-UD-Q3_K_XL-Qwen3.5-4B",
|
| 409 |
hfFilePath: "Qwen3.5-4B-UD-Q3_K_XL-00001-of-00005.gguf",
|
| 410 |
fileSizeInMegabytes: 2436,
|
|
|
|
| 411 |
},
|
| 412 |
"gemma-4-e2b": {
|
| 413 |
...createDefaultModelConfig(),
|
|
|
|
| 415 |
hfRepoId: "Felladrin/gguf-sharded-Q3_K_S-gemma-4-e2b-it",
|
| 416 |
hfFilePath: "gemma-4-e2b-it-Q3_K_S-00001-of-00005.gguf",
|
| 417 |
fileSizeInMegabytes: 2446,
|
|
|
|
| 418 |
},
|
| 419 |
"phi-4-mini-reasoning": {
|
| 420 |
...createDefaultModelConfig(),
|
|
|
|
| 429 |
hfRepoId: "Felladrin/gguf-sharded-Q4_K_M-Spark-X2.6-4B",
|
| 430 |
hfFilePath: "Spark-X2.6-4B-Q4_K_M-00001-of-00007.gguf",
|
| 431 |
fileSizeInMegabytes: 3543,
|
|
|
|
| 432 |
},
|
| 433 |
};
|
docs/ai-integration.md
CHANGED
|
@@ -38,7 +38,7 @@ Two constraints decide what can join that list, both measured against the shippe
|
|
| 38 |
- **Every model file must be a shard.** Chromium's OPFS storage silently truncates files past 2000 MiB, so a monolithic GGUF above that size arrives corrupted in the browser. Shards also keep progressive download working.
|
| 39 |
- **The wasm linear memory ceiling is 4 GiB.** Models at ~3.5 GB and above OOM during buffer reserve, so the practical catalog ceiling is around 3.7 GB of weights.
|
| 40 |
|
| 41 |
-
Models whose chat template thinks by default (Granite 4.2, Gemma 4, Spark X2.6, Qwen 3.5)
|
| 42 |
|
| 43 |
`node scripts/wllama-catalog-check.cjs` HEADs every shard of every catalog model on HuggingFace and verifies the claimed sizes; run it after adding or re-sharding a model.
|
| 44 |
|
|
|
|
| 38 |
- **Every model file must be a shard.** Chromium's OPFS storage silently truncates files past 2000 MiB, so a monolithic GGUF above that size arrives corrupted in the browser. Shards also keep progressive download working.
|
| 39 |
- **The wasm linear memory ceiling is 4 GiB.** Models at ~3.5 GB and above OOM during buffer reserve, so the practical catalog ceiling is around 3.7 GB of weights.
|
| 40 |
|
| 41 |
+
Models whose chat template thinks by default (Granite 4.2, Gemma 4, Spark X2.6, Qwen 3.5) keep thinking: wllama splits their chain of thought into a `reasoning_content` channel, and MiniSearch wraps it in the configured reasoning markers so the thinking UI shows it while the answer follows. Thinking spends the same token budget as the answer, so these models are slower and, on a long chain of thought, can use up the context before answering.
|
| 42 |
|
| 43 |
`node scripts/wllama-catalog-check.cjs` HEADs every shard of every catalog model on HuggingFace and verifies the claimed sizes; run it after adding or re-sharding a model.
|
| 44 |
|
docs/configuration.md
CHANGED
|
@@ -199,7 +199,7 @@ Search results:
|
|
| 199 |
- `{{dateTime}}`: Alias for `{{currentDate}}` - both are replaced with the same current date value
|
| 200 |
- `{{searchResults}}`: Formatted search results from the web search, including the page excerpts when `enablePageContentFetch` is on
|
| 201 |
|
| 202 |
-
**Reasoning Markers:** Models that output internal thought processes use `<think>` and `</think>` markers. The UI extracts and separately displays reasoning content from the final response.
|
| 203 |
|
| 204 |
### Privacy Settings
|
| 205 |
|
|
|
|
| 199 |
- `{{dateTime}}`: Alias for `{{currentDate}}` - both are replaced with the same current date value
|
| 200 |
- `{{searchResults}}`: Formatted search results from the web search, including the page excerpts when `enablePageContentFetch` is on
|
| 201 |
|
| 202 |
+
**Reasoning Markers:** Models that output internal thought processes use `<think>` and `</think>` markers. The UI extracts and separately displays reasoning content from the final response. Browser (Wllama) thinking models are wrapped in these markers automatically, so their chain of thought shows up the same way.
|
| 203 |
|
| 204 |
### Privacy Settings
|
| 205 |
|