Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@DeepSeek FR MCP Server show me the available models"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
DeepSeek FR | DeepSeek V3 & R1 workflows in French
A Model Context Protocol server that exposes the canonical DeepSeek FR knowledge surface — models, prompts, and chat workflows, pricing, FAQ, official links — to MCP-compatible AI clients such as Claude Desktop, Cursor, Windsurf, and Continue. Read-only, no API keys, no quota, ~50 ms cold start.
Official website: https://deepseekfr.org
DeepSeek FR is a French-localized chat interface built around DeepSeek's latest AI models, offering French-speaking users direct access to a capable large language model without needing to navigate an English-first platform. The site provides two model variants — a Pro version aimed at complex reasoning and advanced code tasks, and a Flash version tuned for faster everyday responses — both sharing a one-million-token context window. Users interact through a clean chat interface that supports adjustable reasoning depth, making it practical for everything from quick questions to demanding multi-step analysis. The platform also documents API compatibility with OpenAI and Anthropic conventions, lowering the barrier for developers already familiar with those ecosystems.
Related MCP server: Flux-2 Online MCP
Two model tiers: Pro for deep reasoning and code-heavy work; Flash for lower-latency, everyday use — both with a 1M token context window.
Three reasoning modes: Non-think (maximum speed), Think High (balanced precision), and Think Max (deepest analysis), selectable per conversation.
Extended context handling: Processes long documents such as contracts, research papers, and technical manuals in a single session without chunking.
API compatibility: Works with OpenAI- and Anthropic-style API clients, so existing tooling can connect with minimal changes.
Benchmark transparency: The site publishes performance data across knowledge, reasoning, code, math, and agent workflow benchmarks so users can evaluate fit before integrating.
Local execution support: Documentation covers running models locally with encoding specifications for teams that prefer on-premise deployments.
Document review: Feeding long contracts, compliance documents, or technical manuals into the chat to extract summaries, flag inconsistencies, or answer specific questions.
Code assistance: Reviewing, debugging, or refactoring code across common languages, with the Pro model handling complex architectural questions.
Research and reasoning: Working through multi-step analytical problems — scientific literature, legal arguments, or financial modeling — where depth matters more than speed.
Content drafting: Generating French-language copy, reports, or structured documents for professional or editorial use.
API prototyping: Testing DeepSeek model responses through a familiar chat interface before wiring them into an application via the compatible API.
DeepSeek FR is aimed primarily at French-speaking professionals and developers who want a capable AI assistant without switching to an English-language interface. Developers benefit from the API compatibility and local execution options; researchers and analysts value the long context window and adjustable reasoning depth; writers and content teams find the French-native interface reduces friction in day-to-day drafting tasks. The dual-model approach also makes it accessible to non-technical users who need quick answers as well as to technical teams running sustained, resource-intensive workflows — without requiring them to choose a single tool for both.
list_modelsReturn the canonical list of chat models exposed on the site, with capability notes. (DeepSeek FR)
Input: no parameters. Returns: text/markdown.
get_pricingReturn the canonical pricing entry point for DeepSeek FR.
Input: no parameters. Returns: text/markdown.
get_official_linksReturn the canonical list of official links for DeepSeek FR (website, support, docs when available).
Input: no parameters. Returns: text/markdown.
site://deepseekfr/models — Supported chat models and capability notes.
site://deepseekfr/pricing — Canonical pricing entry point.
site://deepseekfr/faq — Short FAQ generated from public site metadata.
site://deepseekfr/links — Canonical URLs to share with users.
tell_me_about_deepseekfrSummarize what the site is, who it's for, and how it works. — DeepSeek FR
start_chat_session_deepseekfrOpen a chat-evaluation session against the site's models, with sensible defaults. — DeepSeek FR
npx -y @smithery/cli install deepseekfr-mcp --client claude(Replace claude with cursor, windsurf, or continue for those clients.)
git clone https://github.com/rocnubie/deepseekfr-mcp.git
cd deepseekfr-mcp
pnpm installThen add to your MCP client config (claude_desktop_config.json for Claude Desktop, mcp.json for Cursor / Windsurf / Continue):
{
"mcpServers": {
"deepseekfr-mcp": {
"command": "node",
"args": [
"/absolute/path/to/deepseekfr-mcp/src/index.mjs"
]
}
}
}npx @modelcontextprotocol/inspector node src/index.mjsWebsite: https://deepseekfr.org
Pricing: https://deepseekfr.org/pricing
Support: support@deepseekfr.org
pnpm install
pnpm start # run the server over stdioMIT
Return the canonical list of official links for DeepSeek FR (website, support, docs when available).
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It provides useful behavioral context by noting that links are included 'when available', which hints at conditional inclusion. It does not mention side effects or return format, but for a getter with no parameters, this is sufficient and adds value beyond the bare operation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, front-loaded sentence with no wasted words. It states the action ('Return'), the object ('canonical list'), the scope ('DeepSeek FR'), and the details ('website, support, docs when available') efficiently.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given there is no output schema, the description appropriately describes the return value and its contents. It covers the types of links and the 'when available' nuance, which is enough for a 0-parameter tool. It could mention error behavior or empty results, but for this simple tool, it is adequately complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has zero properties, so there are no parameters to explain. Per the rubric, a baseline of 4 applies, and the description does not need to add parameter semantics. It correctly focuses on the output.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool returns the canonical list of official links for DeepSeek FR, specifying the types of links (website, support, docs). The verb 'Return' and resource 'canonical list of official links' make it distinct from sibling tools like list_models and get_pricing, which have different purposes.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description makes it clear when to use the tool (when official links are needed), and the siblings are about models and pricing, so the context is evident. However, it does not explicitly mention alternatives or exclusions, though these are implied by the tool's narrow scope.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Return the canonical pricing entry point for DeepSeek FR.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden for behavioral disclosure. It only states 'Return', implying a read-only operation, but does not describe what the actual return value looks like (e.g., a URL, an object, a text string), any authentication requirements, or potential error behaviors. This lack of detail limits an agent's ability to anticipate outcomes.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, focused sentence with no filler. It is front-loaded with the key verb and object, making it easy to scan.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is extremely simple (no parameters, no output schema), so the minimal description is partially acceptable. However, the phrase 'pricing entry point' remains ambiguous—whether it returns a URL, a data structure, or a human-readable string. Given siblings that are also retrieval tools, more specificity would help complete the picture.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, and schema coverage is 100% (an empty schema). Per the rubric, a baseline of 4 applies. The description adds no parameter information, but with no parameters to document, there is no deficit.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly identifies the tool's action ('Return') and resource ('canonical pricing entry point for DeepSeek FR'). It distinguishes from siblings like list_models and get_official_links by focusing on pricing, though 'entry point' is somewhat abstract and could be more specific (e.g., 'URL' or 'data endpoint').
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance is provided on when to use this tool versus alternatives. The description does not mention any prerequisites, exclusions, or comparisons to sibling tools. For a retrieval tool, this absence is a clear gap.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Return the canonical list of chat models exposed on the site, with capability notes. (DeepSeek FR)
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description indicates a safe read operation with 'Return' and clarifies the scope as 'chat models exposed on the site'. However, with no annotations, it does not disclose potential edge cases, update frequency, or error behavior, and the parenthetical '(DeepSeek FR)' adds ambiguity rather than clarity.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, front-loaded sentence that efficiently states the action and resource. The parenthetical note is minor and does not detract from the density of useful information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a zero-parameter list tool, the description adequately covers the purpose and scope. While it lacks details on the return structure or whether the list is cached or real-time, this is not critical for a straightforward listing operation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, and the input schema is empty. Baseline for no parameters is 4, and the description adds no parameter-related information, which is appropriate since none exist.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool returns the canonical list of chat models with capability notes. It uses a specific verb ('Return') and resource ('canonical list of chat models'), and distinguishes itself from sibling tools (get_pricing, get_official_links) by focusing on model listings.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It is purely declarative, with no mention of use cases, prerequisites, or exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v0.1.0get_official_linksget_pricinglist_modelsEach tool targets a distinct resource: models, pricing, and official links. There is no overlap or ambiguity in their purposes.
All names follow a verb_noun pattern, but list_models uses 'list' while the other two use 'get', creating a minor inconsistency. The pattern is still predictable and readable.
Three tools is well-scoped for a focused reference server that provides canonical information about DeepSeek FR. Each tool serves a clear purpose without redundancy.
The set covers the essential information points: chat models, pricing, and official links. No obvious gaps exist for the server's stated purpose of providing canonical DeepSeek FR references.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Read-only MCP server for Flamel.ai's public content: company overview, blog, case studies, FAQs.
Mitosis agent-memory platform: pricing, docs search, platform status, agent skills. No auth.
Public, read-only MCP server for FarmNeural company facts, packages, and capabilities.
Read-only discovery of Framekeep capabilities, security boundaries, and public resources.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/rocnubie/deepseekfr-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
