llmfit

Default TUI mode and global flags

Synopsis

llmfit [OPTIONS]

Description

The llmfit command launches an interactive Terminal User Interface (TUI) by default. The TUI provides a visual interface for exploring LLM models that fit your system’s hardware specifications.

You can use global flags to customize the behavior or switch to classic CLI table output mode.

Global Flags

These flags work across all subcommands:

--json · boolean · default: "false"

Output results as JSON for tool integration. Works with all subcommands.

--memory · string

Override GPU VRAM size (e.g., “32G”, “32000M”, “1.5T”). Useful when GPU memory autodetection fails.

--max-context · integer

Cap context length used for memory estimation (tokens). Must be >= 1. Falls back to OLLAMA_CONTEXT_LENGTH environment variable if not set.

TUI Mode Flags

These flags apply when running the default TUI:

--cli · boolean · default: "false"

Use classic CLI table output instead of TUI. Shows the fit results in a table format and exits.

-p, --perfect · boolean · default: "false"

Show only models that perfectly match recommended specs.

-n, --limit · integer

Limit number of results displayed.

--sort · enum · default: "score"

Sort column for CLI fit output. Options:

• score - Composite ranking score (default)
• tps - Estimated tokens/second (aliases: tokens, toks, throughput)
• params - Model parameter count
• mem - Memory utilization percentage (aliases: memory, mem_pct, utilization)
• ctx - Context window length (alias: context)
• date - Release date, newest first (aliases: release, released)
• use - Use-case grouping (aliases: use_case, usecase)

Usage Examples

Launch TUI (Default)

# Launch interactive TUI
llmfit

CLI Table Output

# Show fit results as a table
llmfit --cli

# Show only perfect matches
llmfit --cli --perfect

# Show top 10 models sorted by tokens/second
llmfit --cli -n 10 --sort tps

Override GPU Memory

# Override detected VRAM (useful for cloud instances)
llmfit --memory 24G

# Specify memory in different units
llmfit --memory 24576M
llmfit --memory 0.024T

Context Length Control

# Cap context at 8K tokens for memory estimation
llmfit --max-context 8192

# Or set via environment variable
export OLLAMA_CONTEXT_LENGTH=8192
llmfit

JSON Output

# Get JSON output for scripting
llmfit --cli --json -n 5

Example Output

TUI Mode

Launching llmfit displays an interactive interface:

┌─ System ──────────────────────────────────────────────────────────────┐
│ RAM: 64.0 GB  CPU: 16 cores (Apple M2 Max)  GPU: Metal (64.0 GB)     │
└───────────────────────────────────────────────────────────────────────┘

┌─ Filters ─────────────────────────────────────────────────────────────┐
│ Search: [_______]  [A]ll [P]erfect [G]ood [M]arginal                  │
└───────────────────────────────────────────────────────────────────────┘

┌─ Models ──────────────────────────────────────────────────────────────┐
│ Status  Model              Provider   Size   Score  tok/s  Runtime    │
│ ✓ PERF  llama-3.3-70b     Meta       70B    95.2   42.5   MLX        │
│ ✓ PERF  qwen-2.5-72b      Alibaba    72B    94.8   40.1   MLX        │
│ ✓ GOOD  deepseek-v3       DeepSeek   671B   92.1   28.3   llama.cpp  │
└───────────────────────────────────────────────────────────────────────┘

[↑↓] Navigate  [Enter] Details  [/] Search  [q] Quit

CLI Table Output

Running llmfit --cli displays:

╭────────────┬───────────────────┬──────────┬──────┬───────┬──────────────┬─────────┬────────────┬─────────┬────────┬─────────╮
│ Status     │ Model             │ Provider │ Size │ Score │ tok/s est.   │ Quant   │ Runtime    │ Mode    │ Mem %  │ Context │
├────────────┼───────────────────┼──────────┼──────┼───────┼──────────────┼─────────┼────────────┼─────────┼────────┼─────────┤
│ ✓ Perfect  │ llama-3.3-70b     │ Meta     │ 70B  │ 95    │ 42.5         │ 4bit    │ MLX        │ GPU     │ 68.2%  │ 128k    │
│ ✓ Perfect  │ qwen-2.5-72b      │ Alibaba  │ 72B  │ 95    │ 40.1         │ 4bit    │ MLX        │ GPU     │ 71.5%  │ 32k     │
│ ✓ Good     │ deepseek-v3       │ DeepSeek │ 671B │ 92    │ 28.3         │ Q4_K_M  │ llama.cpp  │ GPU     │ 89.7%  │ 128k    │
╰────────────┴───────────────────┴──────────┴──────┴───────┴──────────────┴─────────┴────────────┴─────────┴────────┴─────────╯

Related Commands

• system - Show system hardware specifications
• fit - Find models that fit your system
• recommend - Get top model recommendations
• search - Search for specific models