Skip to content

Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message search, Code Interpreter, langchain, DALL-E-3, OpenAPI Actions, Functions, Secure Multi-User Auth, Presets, open-source for self-hosting. Active

🤖 feat: Tool Calling Support for DeepSeek V3.2 + OpenRouter Reasoning (#10752)

Danny Avila4202db1c9926
diff --git a/api/models/tx.js b/api/models/tx.js
index 4ea93e1a1e..aa213d3475 100644
--- a/api/models/tx.js
+++ b/api/models/tx.js
@@ -141,6 +141,7 @@ const tokenValues = Object.assign(
'command-r': { prompt: 0.5, completion: 1.5 },
'command-r-plus': { prompt: 3, completion: 15 },
'command-text': { prompt: 1.5, completion: 2.0 },
+ 'deepseek-chat': { prompt: 0.28, completion: 0.42 },
'deepseek-reasoner': { prompt: 0.28, completion: 0.42 },
'deepseek-r1': { prompt: 0.4, completion: 2.0 },
'deepseek-v3': { prompt: 0.2, completion: 0.8 },
@@ -246,6 +247,10 @@ const cacheTokenValues = {
'claude-sonnet-4': { write: 3.75, read: 0.3 },
'claude-opus-4': { write: 18.75, read: 1.5 },
'claude-opus-4-5': { write: 6.25, read: 0.5 },
+ // DeepSeek models - cache hit: $0.028/1M, cache miss: $0.28/1M
+ deepseek: { write: 0.28, read: 0.028 },
+ 'deepseek-chat': { write: 0.28, read: 0.028 },
+ 'deepseek-reasoner': { write: 0.28, read: 0.028 },
};
/**
diff --git a/api/models/tx.spec.js b/api/models/tx.spec.js
index 75a8fa1922..18030abb21 100644
--- a/api/models/tx.spec.js
+++ b/api/models/tx.spec.js
@@ -766,6 +766,78 @@ describe('Deepseek Model Tests', () => {
const result = tokenValues[valueKey].prompt && multiplier === tokenValues[valueKey].prompt;
expect(result).toBe(true);
});
+
+ it('should return correct pricing for deepseek-chat', () => {
+ expect(getMultiplier({ model: 'deepseek-chat', tokenType: 'prompt' })).toBe(
+ tokenValues['deepseek-chat'].prompt,
+ );
+ expect(getMultiplier({ model: 'deepseek-chat', tokenType: 'completion' })).toBe(
+ tokenValues['deepseek-chat'].completion,
+ );
+ expect(tokenValues['deepseek-chat'].prompt).toBe(0.28);
+ expect(tokenValues['deepseek-chat'].completion).toBe(0.42);
+ });
+
+ it('should return correct pricing for deepseek-reasoner', () => {
+ expect(getMultiplier({ model: 'deepseek-reasoner', tokenType: 'prompt' })).toBe(
+ tokenValues['deepseek-reasoner'].prompt,
+ );
+ expect(getMultiplier({ model: 'deepseek-reasoner', tokenType: 'completion' })).toBe(
+ tokenValues['deepseek-reasoner'].completion,
+ );
+ expect(tokenValues['deepseek-reasoner'].prompt).toBe(0.28);
+ expect(tokenValues['deepseek-reasoner'].completion).toBe(0.42);
+ });
+
+ it('should handle DeepSeek model name variations with provider prefixes', () => {
+ const modelVariations = [
+ 'deepseek/deepseek-chat',
+ 'openrouter/deepseek-chat',
+ 'deepseek/deepseek-reasoner',
+ ];
+
+ modelVariations.forEach((model) => {
+ const promptMultiplier = getMultiplier({ model, tokenType: 'prompt' });
+ const completionMultiplier = getMultiplier({ model, tokenType: 'completion' });
+ expect(promptMultiplier).toBe(0.28);
+ expect(completionMultiplier).toBe(0.42);
+ });
+ });
+
+ it('should return correct cache multipliers for DeepSeek models', () => {
+ expect(getCacheMultiplier({ model: 'deepseek-chat', cacheType: 'write' })).toBe(
+ cacheTokenValues['deepseek-chat'].write,
+ );
+ expect(getCacheMultiplier({ model: 'deepseek-chat', cacheType: 'read' })).toBe(
+ cacheTokenValues['deepseek-chat'].read,
+ );
+ expect(getCacheMultiplier({ model: 'deepseek-reasoner', cacheType: 'write' })).toBe(
+ cacheTokenValues['deepseek-reasoner'].write,
+ );
+ expect(getCacheMultiplier({ model: 'deepseek-reasoner', cacheType: 'read' })).toBe(
+ cacheTokenValues['deepseek-reasoner'].read,
+ );
+ });
+
+ it('should return correct cache pricing values for DeepSeek models', () => {
+ expect(cacheTokenValues['deepseek-chat'].write).toBe(0.28);
+ expect(cacheTokenValues['deepseek-chat'].read).toBe(0.028);
+ expect(cacheTokenValues['deepseek-reasoner'].write).toBe(0.28);
+ expect(cacheTokenValues['deepseek-reasoner'].read).toBe(0.028);
+ expect(cacheTokenValues['deepseek'].write).toBe(0.28);
+ expect(cacheTokenValues['deepseek'].read).toBe(0.028);
+ });
+
+ it('should handle DeepSeek cache multipliers with model variations', () => {
+ const modelVariations = ['deepseek/deepseek-chat', 'openrouter/deepseek-reasoner'];
+
+ modelVariations.forEach((model) => {
+ const writeMultiplier = getCacheMultiplier({ model, cacheType: 'write' });
+ const readMultiplier = getCacheMultiplier({ model, cacheType: 'read' });
+ expect(writeMultiplier).toBe(0.28);
+ expect(readMultiplier).toBe(0.028);
+ });
+ });
});
describe('Qwen3 Model Tests', () => {
diff --git a/api/package.json b/api/package.json
index 69354e1812..0936bede48 100644
--- a/api/package.json
+++ b/api/package.json
@@ -47,7 +47,7 @@
"@langchain/google-genai": "^0.2.13",
"@langchain/google-vertexai": "^0.2.13",
"@langchain/textsplitters": "^0.1.0",
- "@librechat/agents": "^3.0.34",
+ "@librechat/agents": "^3.0.35",
"@librechat/api": "*",
"@librechat/data-schemas": "*",
"@microsoft/microsoft-graph-client": "^3.0.7",
diff --git a/api/utils/tokens.spec.js b/api/utils/tokens.spec.js
index a169c31622..3336a0f82d 100644
--- a/api/utils/tokens.spec.js
+++ b/api/utils/tokens.spec.js
@@ -665,7 +665,7 @@ describe('Meta Models Tests', () => {
test('should match Deepseek model variations', () => {
expect(getModelMaxTokens('deepseek-chat')).toBe(
- maxTokensMap[EModelEndpoint.openAI]['deepseek'],
+ maxTokensMap[EModelEndpoint.openAI]['deepseek-chat'],
);
expect(getModelMaxTokens('deepseek-coder')).toBe(
maxTokensMap[EModelEndpoint.openAI]['deepseek'],
@@ -677,6 +677,20 @@ describe('Meta Models Tests', () => {
maxTokensMap[EModelEndpoint.openAI]['deepseek.r1'],
);
});
+
+ test('should return 128000 context tokens for all DeepSeek models', () => {
+ expect(getModelMaxTokens('deepseek-chat')).toBe(128000);
+ expect(getModelMaxTokens('deepseek-reasoner')).toBe(128000);
+ expect(getModelMaxTokens('deepseek-r1')).toBe(128000);
+ expect(getModelMaxTokens('deepseek-v3')).toBe(128000);
+ expect(getModelMaxTokens('deepseek.r1')).toBe(128000);
+ });
+
+ test('should handle DeepSeek models with provider prefixes', () => {
+ expect(getModelMaxTokens('deepseek/deepseek-chat')).toBe(128000);
+ expect(getModelMaxTokens('openrouter/deepseek-reasoner')).toBe(128000);
+ expect(getModelMaxTokens('openai/deepseek-v3')).toBe(128000);
+ });
});
describe('matchModelName', () => {
@@ -705,11 +719,42 @@ describe('Meta Models Tests', () => {
});
test('should match Deepseek model variations', () => {
- expect(matchModelName('deepseek-chat')).toBe('deepseek');
+ expect(matchModelName('deepseek-chat')).toBe('deepseek-chat');
expect(matchModelName('deepseek-coder')).toBe('deepseek');
});
});
+ describe('DeepSeek Max Output Tokens', () => {
+ const { getModelMaxOutputTokens } = require('@librechat/api');
+
+ test('should return correct max output tokens for deepseek-chat', () => {
+ expect(getModelMaxOutputTokens('deepseek-chat')).toBe(8000);
+ expect(getModelMaxOutputTokens('deepseek-chat', EModelEndpoint.openAI)).toBe(8000);
+ expect(getModelMaxOutputTokens('deepseek-chat', EModelEndpoint.custom)).toBe(8000);
+ });
+
+ test('should return correct max output tokens for deepseek-reasoner', () => {
+ expect(getModelMaxOutputTokens('deepseek-reasoner')).toBe(64000);
+ expect(getModelMaxOutputTokens('deepseek-reasoner', EModelEndpoint.openAI)).toBe(64000);
+ expect(getModelMaxOutputTokens('deepseek-reasoner', EModelEndpoint.custom)).toBe(64000);
+ });
+
+ test('should return correct max output tokens for deepseek-r1', () => {
+ expect(getModelMaxOutputTokens('deepseek-r1')).toBe(64000);
+ expect(getModelMaxOutputTokens('deepseek-r1', EModelEndpoint.openAI)).toBe(64000);
+ });
+
+ test('should return correct max output tokens for deepseek base pattern', () => {
+ expect(getModelMaxOutputTokens('deepseek')).toBe(8000);
+ expect(getModelMaxOutputTokens('deepseek-v3')).toBe(8000);
+ });
+
+ test('should handle DeepSeek models with provider prefixes for max output tokens', () => {
+ expect(getModelMaxOutputTokens('deepseek/deepseek-chat')).toBe(8000);
+ expect(getModelMaxOutputTokens('openrouter/deepseek-reasoner')).toBe(64000);
+ });
+ });
+
describe('processModelData with Meta models', () => {
test('should process Meta model data correctly', () => {
const input = {
diff --git a/package-lock.json b/package-lock.json
index 8145faae7d..c0f359f04b 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -61,7 +61,7 @@
"@langchain/google-genai": "^0.2.13",
"@langchain/google-vertexai": "^0.2.13",
"@langchain/textsplitters": "^0.1.0",
- "@librechat/agents": "^3.0.34",
+ "@librechat/agents": "^3.0.35",
"@librechat/api": "*",
"@librechat/data-schemas": "*",
"@microsoft/microsoft-graph-client": "^3.0.7",
@@ -16289,9 +16289,9 @@
}
},
"node_modules/@librechat/agents": {
- "version": "3.0.34",
- "resolved": "https://registry.npmjs.org/@librechat/agents/-/agents-3.0.34.tgz",
- "integrity": "sha512-sngGZewCp/p8nMhUVJdXNZw74MpehJ1SMvWPxR9euLdOQX2qgR23/c2/5VxpdsR4gXmOSxpdhsc5cin2y6R/mA==",
+ "version": "3.0.35",
+ "resolved": "https://registry.npmjs.org/@librechat/agents/-/agents-3.0.35.tgz",
+ "integrity": "sha512-9KcTin8CtJIsADkcZtyCHwfn9GQ7AIRWTAhNDFtxXsuqmlnytD8bjh0GiRc6uOOQ/Dw8zL/oRcqNDubempwBfg==",
"license": "MIT",
"dependencies": {
"@langchain/anthropic": "^0.3.26",
@@ -46275,7 +46275,7 @@
"@azure/storage-blob": "^12.27.0",
"@keyv/redis": "^4.3.3",
"@langchain/core": "^0.3.79",
- "@librechat/agents": "^3.0.34",
+ "@librechat/agents": "^3.0.35",
"@librechat/data-schemas": "*",
"@modelcontextprotocol/sdk": "^1.21.0",
"axios": "^1.12.1",
diff --git a/packages/api/package.json b/packages/api/package.json
index c248621148..36a290fb57 100644
--- a/packages/api/package.json
+++ b/packages/api/package.json
@@ -84,7 +84,7 @@
"@azure/storage-blob": "^12.27.0",
"@keyv/redis": "^4.3.3",
"@langchain/core": "^0.3.79",
- "@librechat/agents": "^3.0.34",
+ "@librechat/agents": "^3.0.35",
"@librechat/data-schemas": "*",
"@modelcontextprotocol/sdk": "^1.21.0",
"axios": "^1.12.1",
diff --git a/packages/api/src/utils/tokens.ts b/packages/api/src/utils/tokens.ts
index 7dda247cc1..33196ab7dc 100644
--- a/packages/api/src/utils/tokens.ts
+++ b/packages/api/src/utils/tokens.ts
@@ -140,6 +140,7 @@ const anthropicModels = {
const deepseekModels = {
deepseek: 128000,
+ 'deepseek-chat': 128000,
'deepseek-reasoner': 128000,
'deepseek-r1': 128000,
'deepseek-v3': 128000,
@@ -347,11 +348,21 @@ const anthropicMaxOutputs = {
'claude-3-7-sonnet': 128000,
};
+/** Outputs from https://api-docs.deepseek.com/quick_start/pricing */
+const deepseekMaxOutputs = {
+ deepseek: 8000, // deepseek-chat default: 4K, max: 8K
+ 'deepseek-chat': 8000,
+ 'deepseek-reasoner': 64000, // default: 32K, max: 64K
+ 'deepseek-r1': 64000,
+ 'deepseek-v3': 8000,
+ 'deepseek.r1': 64000,
+};
+
export const maxOutputTokensMap = {
[EModelEndpoint.anthropic]: anthropicMaxOutputs,
[EModelEndpoint.azureOpenAI]: modelMaxOutputs,
- [EModelEndpoint.openAI]: modelMaxOutputs,
- [EModelEndpoint.custom]: modelMaxOutputs,
+ [EModelEndpoint.openAI]: { ...modelMaxOutputs, ...deepseekMaxOutputs },
+ [EModelEndpoint.custom]: { ...modelMaxOutputs, ...deepseekMaxOutputs },
};
/**