Occam

fit.occam/occam

This server exposes symbolic regression through two tools: sindy_run for discovering differential equations from time series data in seconds, and pysr_run for finding algebraic relationships via evolutionary search that returns a Pareto front of expressions ranked by complexity versus accuracy. It wraps PySINDy and SymbolicRegression.jl under the hood. You'd reach for this when you have numerical measurements and want governing equations instead of black box models. The free tier handles up to 100 rows and 8 variables, then switches to pay per use via x402 or card. Connects over streamable HTTP to occam.fit/mcp/ and includes a bootstrap uncertainty tool for getting confidence intervals on fitted constants.

tools: 4
transport: HTTP
last updated: Apr 2, 2026

Open website

Install

From registry metadata

$ npx add-mcp https://occam.fit/mcp/ --name occam

Run in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.

Review the command, arguments and environment values before installing. MCP servers run with your local permissions.

Other ways

Claude Code

$ claude mcp add --transport http occam https://occam.fit/mcp/

Run in your terminal. Add --scope user to make it available in every project.

Tools

Verified live · 4 tools

Verified live against the running server on Nov 6, 2026.

feature_requestRequest a feature that Occam doesn't support yet. Use this when you need a capability that Occam doesn't currently offer. Requests are logged and used to prioritize development. Rate limit: 5 requests/hour per IP, 50/hour global — stricter than the compute tools' 10/hour to pr...1 params

Request a feature that Occam doesn't support yet. Use this when you need a capability that Occam doesn't currently offer. Requests are logged and used to prioritize development. Rate limit: 5 requests/hour per IP, 50/hour global — stricter than the compute tools' 10/hour to pr...

Parameters* required

description*string

A short description of the feature you need. Examples: 'LaTeX output for equations', 'support for ODE constraints', 'GPU-accelerated search', 'larger dataset limits'. Helps prioritize development.

sindy_runSparse Identification of Nonlinear Dynamics (SINDy). Recovers governing differential equations (dx/dt = f(x)) from time series data. Returns human-readable sparse expressions. Fast (seconds). For algebraic y = f(x) relationships without time structure, use pysr_run instead. Pr...7 params

Sparse Identification of Nonlinear Dynamics (SINDy). Recovers governing differential equations (dx/dt = f(x)) from time series data. Returns human-readable sparse expressions. Fast (seconds). For algebraic y = f(x) relationships without time structure, use pysr_run instead. Pr...

Parameters* required

data*array

2D array of time series data. Each row is a timestep, each column is a state variable. Minimum 3 rows (finite-difference derivative). Free tier: 100 rows, 8 variables. Paid tier: up to 500,000 rows...

feature_namesvalue

Names for each variable/feature column. Defaults to x0, x1, ...

max_iterinteger

Maximum STLSQ optimizer iterations. Default 20.default: 20

paymentvalue

Payment credential. Accepts either a JSON object or a JSON-encoded string (FastMCP's transport pre-parses strings whose field annotation is non-bare-`str` into objects, so the object form is canoni...

poly_degreeinteger

Polynomial library degree for SINDy candidate functions. Default 2.default: 2

t*array

Timestamps corresponding to each row of data. Length must match row count.

thresholdnumber

STLSQ sparsity threshold. Higher values produce sparser equations. Default 0.1.default: 0.1

pysr_runEvolutionary Symbolic Regression (PySR). Discovers algebraic equations y = f(x1, x2, ...) from feature/target data. Returns a Pareto front ranked by the complexity/accuracy tradeoff. Slower than SINDy (10-60s); searches often terminate early on convergence. For differential eq...11 params

Evolutionary Symbolic Regression (PySR). Discovers algebraic equations y = f(x1, x2, ...) from feature/target data. Returns a Pareto front ranked by the complexity/accuracy tradeoff. Slower than SINDy (10-60s); searches often terminate early on convergence. For differential eq...

Parameters* required

X*array

2D array of input features. Each row is an observation, each column is a feature. Minimum 5 rows. Free tier: 100 rows, 8 features. Paid tier: up to 50,000 rows, 20 features.

binary_operatorsvalue

Allowed binary operators, drawn from the fixed supported set: +, -, *, /, ^. Custom operators are NOT supported. Default: +, -, *, /. Pass [] for none.

feature_namesvalue

Names for each variable/feature column. Defaults to x0, x1, ...

loss_thresholdvalue

Optional early-stop threshold on the best loss found. If set, the search terminates as soon as any Pareto-front member reaches a loss at or below this value, even if the timeout has not been reache...

max_complexityinteger

Maximum expression tree size. Higher allows more complex expressions. Default 20, max 25.default: 20

paymentvalue

populationsinteger

Number of evolutionary populations for the search. Default 15, max 20.default: 15

stall_detectionboolean

When true (default), the server stops the search early if the best loss has not improved by more than 1% during the last third of the time budget. This reclaims compute once the search has converge...default: true

timeout_secondsinteger

Wall clock time limit in seconds. Free tier: max 60. Paid tier: max 300 (5 minutes). Default 60.default: 60

unary_operatorsvalue

Allowed unary operators, drawn from the fixed supported set: sin, cos, tan, exp, log, log2, log10, sqrt, abs, sinh, cosh, tanh. Custom operators (e.g. 'inv(x) = 1/x') are NOT supported — only the n...

y*array

Target values, one per row of X.

pysr_uncertaintyBootstrap confidence intervals for the numeric constants of a frozen expression, plus optional prediction bands on an x-grid. Typical flow: call pysr_run, pick an expression from the response (best_expression or a pareto_front entry), pass it back here with the same dataset to...8 params

Bootstrap confidence intervals for the numeric constants of a frozen expression, plus optional prediction bands on an x-grid. Typical flow: call pysr_run, pick an expression from the response (best_expression or a pareto_front entry), pass it back here with the same dataset to...

Parameters* required

X*array

2D array of input features. Each row is an observation, each column is a feature. Minimum 5 rows. Free tier: 100 rows, 8 features. Paid tier: up to 50,000 rows, 20 features.

alphanumber

Significance level. 0.05 → 95%% CI. Default 0.05.default: 0.05

expression*string

The expression to bootstrap, as returned by pysr_run (`best_expression` or a `pareto_front[i].expression`). Only numeric Float constants are treated as free parameters — integers in the expression...

feature_namesvalue

Names for each variable/feature column. Defaults to x0, x1, ...

n_resamplesinteger

Number of bootstrap resamples. Higher = tighter CIs, more compute. Default 100.default: 100

x_gridvalue

Optional 2D grid of feature values at which to report a prediction band. Must have the same number of columns as X. Omit to skip prediction-band computation.

y*array

Target values, one per row of X.

y_sigmavalue

Optional per-point measurement standard deviations, or a single scalar applied to all points. When supplied, the helper uses parametric bootstrap (y_b = y + Normal(0, y_sigma)); otherwise it uses r...

Details

Website: occam.fit
Category: Data & Analytics
Transport: HTTP
Registry: MCP registry, active
Resources: 4
Prompts: 2
Tools verified: Nov 6, 2026

Related Data & Analytics MCP servers

View all

ESG MCP Serversfreminder/esg-mcp-servers31 MCP tools for ESG data extraction, PDF processing, vector search, and EU regulation analysis.0stars

Coingecko Mcp Serverfriendlygeorge/coingecko-mcp-serverMCP server for CoinGecko cryptocurrency data API0stars

Defillama Mcp Serverfriendlygeorge/defillama-mcp-serverMCP server for DeFi Llama — free DeFi TVL, yields, stablecoins, bridges, DEX, and fee data for AI...0stars

Defillama Mcpglebenjoy/defillama-mcpDefiLlama DeFi data: chain/protocol TVL, yields, stablecoins, token prices. No API key.0stars

Dexscreener Mcpglebenjoy/dexscreener-mcpDexScreener on-chain market data: search pairs, tokens, pools, trending tokens. No API key.0stars

Hyperliquid Info Mcpglebenjoy/hyperliquid-info-mcpHyperliquid perp & spot data: prices, funding, open interest, order book, positions. No API key.0stars