mlx-chronos 0.1.2

Benchmark suite and community leaderboard for local LLM inference on Apple Silicon

mlx-Chronos ⏱️

Benchmark suite and community leaderboard for local LLM inference on Apple Silicon.
Run it. Share your results. Compare across hardware.

---

What is mlx-Chronos?

mlx-Chronos is a standardized benchmarking tool for local LLM inference engines on Apple Silicon. It automatically detects your hardware, runs a consistent set of tests across installed engines, and produces a structured JSON result you can contribute to the community leaderboard.

Supported engines:

• Ollama (MLX backend)
• oMLX
• Rapid-MLX
• mlx-lm (Apple MLX)

Metrics measured:

• TTFT — Time to First Token (cold and cached, with statistics)
• tok/s — Client-observed request throughput (mean, stddev, min, max across trials)
• Sustained tok/s — Optional long throughput profile for heat buildup and late-run degradation checks
• Output tokens — Completion token counts for throughput trials
• System RAM peak — Peak total Mac RAM in use during the benchmark, used as the public memory comparison metric
• Engine RSS — Diagnostic peak RSS of the engine server process when available
• Tool calling — Success rate (coming in v0.2)

---

How It Works

When you run mlx-Chronos, it executes a fixed benchmark protocol against the running engine:

Cold TTFT — sends a prompt to the model and measures the time from request to first non-empty streamed token, including whitespace-only text tokens. Each trial uses a unique prompt to avoid cache hits.

Cached TTFT — sends the same fixed prompt on every cached trial. A priming call loads it into cache first, then cached trials run consecutively. This measures cache performance without interleaving unrelated prompts between cached measurements.

Request throughput (tok/s) — measures completion tokens divided by the full client-observed request time for a standard fixed prompt. This includes request overhead, prefill, and decode, so it is an end-to-end throughput metric rather than pure decode speed. New runs also record client-observed decode_tokens_per_second from the streaming throughput trial when reliable completion-token usage is available. If an engine cannot provide usage in the streaming response, the run falls back to a local estimate and is marked as not leaderboard-comparable. Throughput uses a fixed requested max_tokens value by default, and optional output token bounds can be requested with --max-tokens / --min-tokens.

System RAM peak — continuously samples total Mac RAM usage from before warmup through the recorded benchmark phases and reports the observed peak in GB and percent. This is the public leaderboard memory metric because it answers the practical question of how much memory pressure the run placed on the Mac.

Thermal monitor — records phase timings plus a lightweight thermal summary during the run when macOS thermal state is available through Foundation/PyObjC. New JSON results include start/end/worst thermal state, sample count, and phases where non-nominal thermal state was observed.

Sustained profile — --profile sustained runs a single long throughput trial with max_tokens=1000 by default and records progress samples every 100 generated output units. These samples make late-run throughput drops easier to spot. When a sustained run also observes a thermal-state change or non-nominal thermal state, mlx-Chronos records a sustained throttling warning in result metadata.

Cooldown tracking — before each run, mlx-Chronos checks the latest prior JSON result in the same output directory. The elapsed time is saved as meta.elapsed_since_last_benchmark_seconds; --cooldown-seconds can enforce a pause before starting a new run.

Peak engine RSS — records the resident memory of the engine server process after warmup, through the recorded benchmark phases, when the process can be identified. This is diagnostic only: it is not total model memory or a public efficiency ranking metric, because macOS/Metal unified-memory accounting can vary across environments. The default RAM sampling interval is 50ms and can be changed with --ram-sample-interval.

All metrics are run over multiple trials and reported with mean, stddev, min, and max. p95 is added only when at least 20 trials are available. The default is 5 trials, with a maximum of 30 unique cold prompts. Results are saved as structured JSON in results/local/ by default. Maintainers publish reviewed JSON files into results/submitted/ after accepting them for the community leaderboard. New result JSON also records the benchmark protocol metadata, including exact prompt text and requested token bounds, so runs can be reproduced without digging through source code. Current protocol v2 throughput uses streaming requests with usage metadata. Older protocol v1 results used non-streaming throughput, so compare those rows with that workload difference in mind.

---

Community Leaderboard

View the full leaderboard with all submitted results:

→ igurss.github.io/mlx-chronos

The leaderboard supports model search plus engine, chip, machine model, memory, and throughput max-token filters so contributors can quickly compare a specific model across local inference engines and Apple Silicon hardware.

---

Current Release

0.1.2 is a compatibility-preserving patch release over 0.1.1. It adds sustained benchmark runs, cooldown metadata, throughput token-bound metadata, client-observed decode throughput when reliable usage data is available, and clearer warnings for fallback token counts and unknown engine versions.

---

Quick Start

# Install
pip install mlx-chronos

# Check available engines
mlx-chronos engines

# Validate setup before a run
mlx-chronos validate --engine omlx --model "Qwen3.5-4B-OptiQ-4bit"

# Run benchmark (JSON by default)
mlx-chronos run --engine omlx --model "Qwen3.5-4B-OptiQ-4bit"

# Optional: request throughput output token bounds
mlx-chronos run --engine omlx --model "Qwen3.5-4B-OptiQ-4bit" --max-tokens 100 --min-tokens 80

# Optional: sustained profile for a longer heat/throttling-sensitive run
mlx-chronos run --engine omlx --model "Qwen3.5-4B-OptiQ-4bit" --profile sustained

# Optional: enforce a pause after a recent run in the same output directory
mlx-chronos run --engine omlx --model "Qwen3.5-4B-OptiQ-4bit" --cooldown-seconds 300

# Optional: write both JSON and Markdown outputs
mlx-chronos run --engine omlx --model "Qwen3.5-4B-OptiQ-4bit" --format all

# Optional: choose a custom output directory
mlx-chronos run --engine omlx --model "Qwen3.5-4B-OptiQ-4bit" --output-dir ~/Desktop/benchmarks

Note: the engine server must be running before you launch mlx-chronos. See CONTRIBUTING.md for setup instructions.

---

Contributing Your Results

1. Run mlx-chronos run on your Mac
2. A JSON file is generated in results/local/ (use --format all for a Markdown summary too)
3. Check the result without sending it:

   mlx-chronos submit --file results/local/your-result.json --dry-run
   

4. Send the JSON to the maintainer inbox:

   mlx-chronos submit --file results/local/your-result.json
   

5. The maintainer reviews accepted JSON files and publishes verified results manually

Leaderboard submissions must report throughput using the engine response's usage.completion_tokens. Local runs can still be saved with a fallback token estimate, but those results are not accepted for the public leaderboard and are marked with meta.word_fallback_warning.

Maintainers can override the public inbox endpoint with --endpoint or the MLX_CHRONOS_SUBMIT_ENDPOINT environment variable. The command sends the JSON file as result_json plus brief form metadata so the inbox provider does not classify the submission as blank spam. To include a real contact address, pass --email or set MLX_CHRONOS_SUBMITTER_EMAIL.

See CONTRIBUTING.md for detailed instructions.

---

Benchmark Methodology

See docs/methodology.md for a full explanation of what is measured, how, and why.

---

Roadmap

Completed

• Core benchmark runner with repeated trials, warmup, cache priming, and phase-separated metrics
• Engine support for oMLX, Rapid-MLX, mlx-lm, and Ollama
• Hardware detection for chip, machine model, memory, macOS, Python, architecture, and thermal state
• Strict JSON schema validation with raw-trial consistency checks
• Continuous engine RSS and system RAM peak sampling
• Preflight validation for engine, server, and model access
• GitHub Actions validation for submitted results
• GitHub Pages leaderboard with model search and engine/chip/machine/memory filters
• JSON and Markdown result export
• mlx-chronos submit for sending validated JSON results to the maintainer inbox
• Published Apple M2 sample results refreshed with the current benchmark protocol
• Warnings for battery mode, Low Power Mode, non-nominal thermal state, and unavailable thermal state
• Integration tests against mock OpenAI-compatible servers
• Larger fixed cold-prompt pool with optional p95 reporting for larger runs
• Request-throughput timing metadata and client-observed streaming decode throughput
• Phase timing metadata and lightweight continuous thermal monitoring
• Sustained benchmark profile, cooldown metadata, and max-token leaderboard filter

Next

• Add richer benchmark condition metadata without breaking the v0.1 JSON contract

Future

• Evaluate a clearer TTFT naming model without breaking the v0.1 JSON contract
• Add tool-calling success-rate benchmarks
• Explore anti-spoofing checks for community submissions
• Document external contributor branch workflow when community PRs start arriving
• Collect more results from M3, M4, and M5 systems

---

License

Apache 2.0 — see LICENSE