Open-source AI browser agent for Chrome and Firefox (monorepo) 🧠
JavaScript ai-agentai-agentsai-toolsbrowser-aibrowser-automationbrowser-extensionchrome-aichrome-devtoolschrome-extensionchrome-extensionsfirefox-addonfirefox-addonsfirefox-extensionfirefox-extensionsgemmagemma4minimaxoffline-aioffline-llmqwen
feat(providers): recognize DeepSeek V4.1-Flash-1017 snapshot ids
Codex release-state test40dc709ba574
diff --git a/docs/fr/providers-and-models.md b/docs/fr/providers-and-models.mdindex e99eb682f..41fde5c77 100644--- a/docs/fr/providers-and-models.md+++ b/docs/fr/providers-and-models.md@@ -235,7 +235,7 @@ Le mode Ask ignore le niveau du fournisseur et reste en lecture seule. Le mode A| Fournisseur | Mécanisme ||---|---|| Compatible OpenAI | Regex sur le nom du modèle (`gpt-4o`, `gpt-5`, `claude-3`, `claude-sonnet-4`, `gemini-2.0-flash`, etc.) |-| DeepSeek | La famille `deepseek-flash` (y compris les alias retirés `deepseek-v4-flash`) est multimodale ; `deepseek-v4-pro` et les identifiants de la génération V3 sont textuels |+| DeepSeek | La famille `deepseek-flash` (y compris les alias retirés `deepseek-v4-flash` et les instantanés datés des routeurs comme `deepseek-v4.1-flash-1017`) est multimodale ; `deepseek-v4-pro` et les identifiants de la génération V3 sont textuels || Anthropic | Patterns `claude-(3\|sonnet-4\|opus-4)` || Ollama | `POST /api/show` `capabilities`, avec replis historiques `projector_info` / `.vision.` || llama.cpp | `GET /props` → `modalities.vision`, avec Automatique / Forcer / Désactivé |@@ -261,12 +261,15 @@ Lorsque le fournisseur actif est Anthropic, l'agent convertit les messages au fo### DeepSeek-Le modèle livré est `deepseek-flash` (DeepSeek-V4.1-Flash). Les identifiants retirés-`deepseek-v4-flash` et `deepseek-v4-flash-vision-exp` sont toujours servis par le même-modèle et facturés comme Flash : ils conservent donc le support complet (contexte 1M,-sortie 384K, entrée d'images). Tous les autres identifiants DeepSeek — y compris-`deepseek-v4-pro`, retiré — restent sur un profil prudent (contexte 64K, sortie 8K,-texte uniquement) au lieu d'hériter de capacités qu'ils n'ont peut-être pas.+Le modèle livré est `deepseek-flash` (DeepSeek-V4.1-Flash, actualisé le 17 octobre+2026 vers l'instantané `-1017`, sans changement d'API ni de tarif). Les identifiants+retirés `deepseek-v4-flash` et `deepseek-v4-flash-vision-exp` sont toujours servis par+le même modèle et facturés comme Flash, et les instantanés datés des routeurs tels que+`deepseek/deepseek-v4.1-flash-1017` (OpenRouter) y renvoient également : ils conservent+donc le support complet (contexte 1M, sortie 384K, entrée d'images). Tous les autres+identifiants DeepSeek — y compris `deepseek-v4-pro`, retiré — restent sur un profil+prudent (contexte 64K, sortie 8K, texte uniquement) au lieu d'hériter de capacités+qu'ils n'ont peut-être pas.| Aspect | Comportement ||---|---|diff --git a/docs/providers-and-models.md b/docs/providers-and-models.mdindex e7b5aa76c..cb1a48e91 100644--- a/docs/providers-and-models.md+++ b/docs/providers-and-models.md@@ -670,7 +670,7 @@ Ask mode ignores provider tier and stays read-only. Act mode uses the selected t| Provider | Mechanism || ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- || OpenAI-compatible | Regex against model name (`gpt-4o`, `gpt-5`, `claude-3`, `claude-sonnet-4`, `gemini-2.0-flash`, etc.) |-| DeepSeek | The `deepseek-flash` family (including the retired `deepseek-v4-flash` aliases) is multimodal; `deepseek-v4-pro` and the V3-era ids are text-only |+| DeepSeek | The `deepseek-flash` family (including the retired `deepseek-v4-flash` aliases and dated router snapshot ids such as `deepseek-v4.1-flash-1017`) is multimodal; `deepseek-v4-pro` and the V3-era ids are text-only || Anthropic | `claude-(3\|sonnet-4\|opus-4)` patterns || Ollama | `POST /api/show` `capabilities`, with legacy projector / `.vision.` metadata fallbacks; Auto / Force on / Off || llama.cpp | `GET /props` → `modalities.vision`, with Auto / Force on / Off |@@ -696,12 +696,14 @@ When the active provider is Anthropic, the agent converts OpenAI-format messages### DeepSeek-`deepseek-flash` (DeepSeek-V4.1-Flash) is the shipped model. The retired+`deepseek-flash` (DeepSeek-V4.1-Flash, refreshed to the `-1017` snapshot on+2026-10-17 with no API or price change) is the shipped model. The retired`deepseek-v4-flash` and `deepseek-v4-flash-vision-exp` ids still serve the same-model and are billed as Flash, so they keep full support (1M context, 384K-output, image input). Every other DeepSeek id — including the retired-`deepseek-v4-pro` — stays on a conservative profile (64K context, 8K output,-text-only) rather than inheriting capacities it may not have.+model and are billed as Flash, and router snapshot ids such as OpenRouter's+`deepseek/deepseek-v4.1-flash-1017` resolve to it too, so they keep full support+(1M context, 384K output, image input). Every other DeepSeek id — including the+retired `deepseek-v4-pro` — stays on a conservative profile (64K context, 8K+output, text-only) rather than inheriting capacities it may not have.| Aspect | Behaviour || ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |diff --git a/docs/zh-CN/providers-and-models.md b/docs/zh-CN/providers-and-models.mdindex 557f2897e..84ac7b030 100644--- a/docs/zh-CN/providers-and-models.md+++ b/docs/zh-CN/providers-and-models.md@@ -203,7 +203,7 @@ Ask 模式忽略提供商层级,保持只读。Act 模式使用所选层级的| 提供商 | 机制 ||---|---|| OpenAI 兼容 | 根据模型名称进行正则匹配(`gpt-4o`、`gpt-5`、`claude-3`、`claude-sonnet-4`、`gemini-2.0-flash` 等) |-| DeepSeek | `deepseek-flash` 系列(含已退役的 `deepseek-v4-flash` 别名)支持多模态;`deepseek-v4-pro` 与 V3 时代标识为纯文本 |+| DeepSeek | `deepseek-flash` 系列(含已退役的 `deepseek-v4-flash` 别名及路由器日期快照,如 `deepseek-v4.1-flash-1017`)支持多模态;`deepseek-v4-pro` 与 V3 时代标识为纯文本 || Anthropic | `claude-(3\|sonnet-4\|opus-4)` 模式 || Ollama | `POST /api/show` 的 `capabilities`,并兼容旧版 `projector_info` / `.vision.` 元数据 || llama.cpp | `GET /props` → `modalities.vision`,支持自动 / 强制开启 / 关闭 |@@ -227,8 +227,10 @@ Ask 模式忽略提供商层级,保持只读。Act 模式使用所选层级的### DeepSeek-出厂模型是 `deepseek-flash`(DeepSeek-V4.1-Flash)。已退役的 `deepseek-v4-flash` 与-`deepseek-v4-flash-vision-exp` 仍由同一模型承接并按 Flash 计费,因此保留完整支持+出厂模型是 `deepseek-flash`(DeepSeek-V4.1-Flash,2026 年 10 月 17 日更新为 `-1017`+快照,API 与价格不变)。已退役的 `deepseek-v4-flash` 与 `deepseek-v4-flash-vision-exp`+仍由同一模型承接并按 Flash 计费;OpenRouter 等路由发布的日期快照+(`deepseek/deepseek-v4.1-flash-1017`)同样指向该模型,因此保留完整支持(1M 上下文、384K 输出、图片输入)。其他 DeepSeek 标识(包括已退役的`deepseek-v4-pro`)保持保守配置(64K 上下文、8K 输出、纯文本),不会继承它未必具备的更大容量。diff --git a/src/chrome/src/providers/deepseek-config.js b/src/chrome/src/providers/deepseek-config.jsindex 930d38b63..e5b58e382 100644--- a/src/chrome/src/providers/deepseek-config.js+++ b/src/chrome/src/providers/deepseek-config.js@@ -19,9 +19,11 @@ export const DEEPSEEK_BASE_URL = 'https://api.deepseek.com';export const DEEPSEEK_LEGACY_DEFAULT_BASE_URL = 'https://api.deepseek.com/v1';export const DEEPSEEK_ANTHROPIC_BASE_URL = 'https://api.deepseek.com/anthropic';-// `deepseek-flash` (DeepSeek-V4.1-Flash) is the current model id. The retired-// `deepseek-v4-flash` / `deepseek-v4-flash-vision-exp` ids still call the same-// model and are billed as Flash, so they keep first-class handling.+// `deepseek-flash` (DeepSeek-V4.1-Flash) is the current model id; DeepSeek has+// served the -1017 snapshot behind it since 2026-10-17 with no API or price+// change. The retired `deepseek-v4-flash` / `deepseek-v4-flash-vision-exp` ids+// still call the same model and are billed as Flash, so they keep first-class+// handling.export const DEEPSEEK_DEFAULT_MODEL = 'deepseek-flash';export const DEEPSEEK_LEGACY_DEFAULT_MODEL = 'deepseek-v4-flash';@@ -47,13 +49,17 @@ export function isDeepSeekModel(model) {}/**- * V4.1-Flash family: the current id plus every retired alias that resolves to- * it. Only this family is multimodal — `deepseek-v4-pro` (DeepSeek-V4-Pro-0813)- * and the V3-era chat/reasoner ids are text-only.+ * V4.1-Flash family: the current id, every retired alias that resolves to it,+ * and the dated snapshot ids routers publish (OpenRouter's+ * `deepseek/deepseek-v4.1-flash-1017`). Only this family is multimodal —+ * `deepseek-v4-pro` (DeepSeek-V4-Pro-0813) and the V3-era chat/reasoner ids are+ * text-only.*/export function isDeepSeekV41FlashModel(model) {const m = clean(model);- return m.includes('deepseek-flash') || m.includes('deepseek-v4-flash');+ return m.includes('deepseek-flash')+ || m.includes('deepseek-v4-flash')+ || m.includes('deepseek-v4.1-flash');}/**diff --git a/src/chrome/src/ui/settings.js b/src/chrome/src/ui/settings.jsindex dd898f6b4..954fe5abd 100644--- a/src/chrome/src/ui/settings.js+++ b/src/chrome/src/ui/settings.js@@ -3353,7 +3353,7 @@ function renderProviders() {fields: [{ key: 'apiKey', labelKey: 'st.provider.field.api_key', type: 'password', placeholder: 'sk-or-...' },{ key: 'model', labelKey: 'st.provider.field.model', type: 'text', placeholder: 'openrouter/free',- suggestions: ['openrouter/free', 'google/gemini-4-argon', 'meta/muse-wonder', 'meta/muse-spark-open', 'anthropic/claude-opus-5.5', 'qwen/qwen3.8-27b', 'moonshotai/kimi-k3', 'z-ai/glm-5.3', 'minimax/minimax-m3'] },+ suggestions: ['openrouter/free', 'deepseek/deepseek-v4.1-flash-1017', 'google/gemini-4-argon', 'meta/muse-wonder', 'meta/muse-spark-open', 'anthropic/claude-opus-5.5', 'qwen/qwen3.8-27b', 'moonshotai/kimi-k3', 'z-ai/glm-5.3', 'minimax/minimax-m3'] },{ key: 'baseUrl', labelKey: 'st.provider.field.api_base_url', type: 'text', placeholder: 'https://openrouter.ai/api/v1' },PROMPT_TIER_FIELD,TOOL_MODE_FIELD,diff --git a/src/firefox/src/providers/deepseek-config.js b/src/firefox/src/providers/deepseek-config.jsindex 930d38b63..e5b58e382 100644--- a/src/firefox/src/providers/deepseek-config.js+++ b/src/firefox/src/providers/deepseek-config.js@@ -19,9 +19,11 @@ export const DEEPSEEK_BASE_URL = 'https://api.deepseek.com';export const DEEPSEEK_LEGACY_DEFAULT_BASE_URL = 'https://api.deepseek.com/v1';export const DEEPSEEK_ANTHROPIC_BASE_URL = 'https://api.deepseek.com/anthropic';-// `deepseek-flash` (DeepSeek-V4.1-Flash) is the current model id. The retired-// `deepseek-v4-flash` / `deepseek-v4-flash-vision-exp` ids still call the same-// model and are billed as Flash, so they keep first-class handling.+// `deepseek-flash` (DeepSeek-V4.1-Flash) is the current model id; DeepSeek has+// served the -1017 snapshot behind it since 2026-10-17 with no API or price+// change. The retired `deepseek-v4-flash` / `deepseek-v4-flash-vision-exp` ids+// still call the same model and are billed as Flash, so they keep first-class+// handling.export const DEEPSEEK_DEFAULT_MODEL = 'deepseek-flash';export const DEEPSEEK_LEGACY_DEFAULT_MODEL = 'deepseek-v4-flash';@@ -47,13 +49,17 @@ export function isDeepSeekModel(model) {}/**- * V4.1-Flash family: the current id plus every retired alias that resolves to- * it. Only this family is multimodal — `deepseek-v4-pro` (DeepSeek-V4-Pro-0813)- * and the V3-era chat/reasoner ids are text-only.+ * V4.1-Flash family: the current id, every retired alias that resolves to it,+ * and the dated snapshot ids routers publish (OpenRouter's+ * `deepseek/deepseek-v4.1-flash-1017`). Only this family is multimodal —+ * `deepseek-v4-pro` (DeepSeek-V4-Pro-0813) and the V3-era chat/reasoner ids are+ * text-only.*/export function isDeepSeekV41FlashModel(model) {const m = clean(model);- return m.includes('deepseek-flash') || m.includes('deepseek-v4-flash');+ return m.includes('deepseek-flash')+ || m.includes('deepseek-v4-flash')+ || m.includes('deepseek-v4.1-flash');}/**diff --git a/src/firefox/src/ui/settings.js b/src/firefox/src/ui/settings.jsindex 514ae28ab..d48165818 100644--- a/src/firefox/src/ui/settings.js+++ b/src/firefox/src/ui/settings.js@@ -2932,7 +2932,7 @@ function renderProviders() {fields: [{ key: 'apiKey', labelKey: 'st.provider.field.api_key', type: 'password', placeholder: 'sk-or-...' },{ key: 'model', labelKey: 'st.provider.field.model', type: 'text', placeholder: 'openrouter/free',- suggestions: ['openrouter/free', 'google/gemini-4-argon', 'meta/muse-wonder', 'meta/muse-spark-open', 'anthropic/claude-opus-5.5', 'qwen/qwen3.8-27b', 'moonshotai/kimi-k3', 'z-ai/glm-5.3', 'minimax/minimax-m3'] },+ suggestions: ['openrouter/free', 'deepseek/deepseek-v4.1-flash-1017', 'google/gemini-4-argon', 'meta/muse-wonder', 'meta/muse-spark-open', 'anthropic/claude-opus-5.5', 'qwen/qwen3.8-27b', 'moonshotai/kimi-k3', 'z-ai/glm-5.3', 'minimax/minimax-m3'] },{ key: 'baseUrl', labelKey: 'st.provider.field.api_base_url', type: 'text', placeholder: 'https://openrouter.ai/api/v1' },PROMPT_TIER_FIELD,TOOL_MODE_FIELD,diff --git a/test/provider-model-limits.mjs b/test/provider-model-limits.mjsindex d37985360..72131b8c1 100644--- a/test/provider-model-limits.mjs+++ b/test/provider-model-limits.mjs@@ -8,7 +8,7 @@ const ROOT = path.resolve(path.dirname(fileURLToPath(import.meta.url)), '..');for (const browser of ['chrome', 'firefox']) {const prefix = path.join(ROOT, 'src', browser, 'src');const { BaseLLMProvider } = await import(pathToFileURL(path.join(prefix, 'providers', 'base.js')).href);- const { inferMaxOutputTokens, resolveMaxOutputTokens } = await import(+ const { inferContextWindow, inferMaxOutputTokens, resolveMaxOutputTokens } = await import(pathToFileURL(path.join(prefix, 'providers', 'context-windows.js')).href);@@ -26,8 +26,11 @@ for (const browser of ['chrome', 'firefox']) {assert.equal(inferMaxOutputTokens({ model: 'gpt-4o' }), 16384, `${browser}: GPT-4o output ceiling`);assert.equal(inferMaxOutputTokens({ model: 'deepseek-v4-flash' }), 384000, `${browser}: DeepSeek V4 output ceiling`);assert.equal(inferMaxOutputTokens({ model: 'deepseek-flash' }), 384000, `${browser}: current DeepSeek output ceiling`);+ assert.equal(inferMaxOutputTokens({ model: 'deepseek/deepseek-v4.1-flash-1017' }), 384000, `${browser}: OpenRouter DeepSeek V4.1 snapshot output ceiling`);assert.equal(inferMaxOutputTokens({ model: 'deepseek-v4-pro' }), 8192, `${browser}: retired DeepSeek Pro stays conservative`);assert.equal(inferMaxOutputTokens({ model: 'deepseek-chat' }), 8192, `${browser}: legacy DeepSeek chat output ceiling`);+ assert.equal(inferContextWindow({ category: 'router', model: 'deepseek/deepseek-v4.1-flash-1017' }), 1000000, `${browser}: OpenRouter DeepSeek V4.1 snapshot context window`);+ assert.equal(inferContextWindow({ category: 'router', model: 'deepseek/deepseek-v4.1-flash' }), 1000000, `${browser}: OpenRouter DeepSeek V4.1 alias context window`);assert.equal(resolveMaxOutputTokens({ model: 'claude-haiku-4-5', maxOutputTokens: 128000 }),64000,diff --git a/test/run.js b/test/run.jsindex 896253531..297116189 100644--- a/test/run.js+++ b/test/run.js@@ -77102,6 +77102,11 @@ test('DeepSeek Chat Completions uses native thinking, vision, streaming, and rep{ contextWindow: 1000000, maxOutputTokens: 384000, vision: true },`${label}: retired flash aliases keep the V4.1-Flash capacities`,);+ assert.deepEqual(+ deepSeekConfig.deepSeekModelCapabilities('deepseek/deepseek-v4.1-flash-1017'),+ { contextWindow: 1000000, maxOutputTokens: 384000, vision: true },+ `${label}: OpenRouter V4.1-Flash snapshots keep the full capacities`,+ );assert.deepEqual(deepSeekConfig.deepSeekModelCapabilities('deepseek-v4-pro'),{ contextWindow: 65536, maxOutputTokens: 8192, vision: false },@@ -77253,6 +77258,23 @@ test('DeepSeek Chat Completions uses native thinking, vision, streaming, and rep`${PM.name}: a router-hosted DeepSeek Flash model uses dedicated capability detection`,);assert.equal(routerFlash.supportsVision, true, `${PM.name}: router-hosted DeepSeek Flash aliases retain vision detection`);+ const routerSnapshot = manager._createProvider('openrouter', {+ type: 'openai',+ category: 'router',+ providerName: 'openrouter',+ baseUrl: 'https://openrouter.ai/api/v1',+ model: 'deepseek/deepseek-v4.1-flash-1017',+ });+ assert.equal(+ routerSnapshot.constructor.name,+ 'DeepSeekProvider',+ `${PM.name}: the OpenRouter V4.1-Flash snapshot uses dedicated capability detection`,+ );+ assert.equal(+ routerSnapshot.supportsVision,+ true,+ `${PM.name}: the OpenRouter V4.1-Flash snapshot stays multimodal`,+ );const migrated = manager._migrateStoredProviderConfigs({deepseek: {model: 'deepseek-v4-flash',