--- license: mit library_name: gguf pipeline_tag: time-series-forecasting language: - en base_model: google/timesfm-2.5-200m-pytorch base_model_relation: quantized quantized_by: amaye15 tags: - gguf - time-series - forecasting - zero-shot - probabilistic - transformer - patch-based - rust inference: false --- # timesfm-rs Pure Rust converter and inference engine for [google/timesfm-2.5-200m-pytorch](https://huggingface.co/google/timesfm-2.5-200m-pytorch). Pre-converted GGUF files are available at [amaye15/timesfm-gguf](https://huggingface.co/amaye15/timesfm-gguf). Produces GGUF v3 files and runs native forecasting — no Python required. ## Build ```bash cargo build --release ``` ## Convert Downloads the model from HuggingFace and writes a GGUF file: ```bash # F16 (recommended) ./target/release/timesfm-rs convert --model google/timesfm-2.5-200m-pytorch --dtype f16 --output gguf/timesfm-f16.gguf # Q8_0 (smallest) ./target/release/timesfm-rs convert --dtype q8 --output gguf/timesfm-q8.gguf # F32 (full precision) ./target/release/timesfm-rs convert --dtype f32 --output gguf/timesfm-f32.gguf ``` To convert all dtypes at once: ```bash ./scripts/convert_all.sh ``` HuggingFace token (optional for public models): ```bash HF_TOKEN=hf_... ./scripts/convert_all.sh ``` ## Inspect tensors Print all tensor names and shapes from a GGUF file: ```bash ./target/release/timesfm-rs inspect-tensors gguf/timesfm-f16.gguf ``` ## Infer Run forecasting from comma-separated context values: ```bash echo '{"context": [1.0, 1.2, 1.5, 1.3, 1.8, 2.0, 1.9, 2.1], "horizon": 64}' \ | ./target/release/timesfm-rs infer --gguf gguf/timesfm-f16.gguf ``` Output is JSON in an OpenAI-compatible forecast format with point forecast and all 9 quantile channels (q0.1–q0.9): ```json { "id": "forecast-000001932b7a1234", "object": "forecast", "created": 1749686400, "model": "timesfm", "choices": [{ "index": 0, "forecast": { "point": [2.1, 2.3, 2.5, "..."], "quantiles": { "0.10": [1.8, 2.0, 2.2, "..."], "0.50": [2.1, 2.3, 2.5, "..."], "0.90": [2.4, 2.6, 2.8, "..."] } }, "finish_reason": "stop" }], "usage": {"context_length": 8, "forecast_length": 64} } ``` **Batch inference** — pass multiple series as a nested array to get one `Choice` per series: ```bash echo '{"context": [[1.0, 1.2, 1.5], [2.0, 2.2, 2.5]], "horizon": 64}' \ | ./target/release/timesfm-rs infer --gguf gguf/timesfm-f16.gguf ``` ## Python bindings Install with [maturin](https://github.com/PyO3/maturin) inside a virtual environment: ```bash python -m venv .venv && source .venv/bin/activate pip install maturin maturin develop --features python ``` ```python import timesfm_rs model = timesfm_rs.TimesFM("gguf/timesfm-f16.gguf") result = model.forecast([1.0, 1.2, 1.5, 1.3, 1.8, 2.0], horizon=64) fc = result["choices"][0]["forecast"] point = fc["point"] # median forecast q10 = fc["quantiles"]["0.10"] # 10th-percentile q90 = fc["quantiles"]["0.90"] # 90th-percentile # Batch — one Choice per series result = model.forecast([[1.0, 1.2, 1.5], [2.0, 2.2, 2.5]], horizon=64) ``` `forecast` returns a Python dict in the same OpenAI-compatible format as the CLI. ## Architecture notes TimesFM 2.5 is a decoder-only patch transformer: - **Input**: Context is split into 32-value patches; each patch is instance-normalized (RevIN with cumulative running statistics) and concatenated with an observation mask, producing 64-dim tokenizer inputs - **Backbone**: 20-layer causal transformer, d_model=1280, 16 heads, head_dim=80, d_ff=1280; 200M parameters - **Attention**: Sandwich norms (pre-norm → attention → post-norm + residual); fused QKV projection; per-dimension query scale `1.442695/√80 × softplus(param)`; QK RMSNorm after RoPE - **AR decoding**: KV cache with 4-patch decode stride — O(n) instead of O(n2) re-forward per step - **Output**: Each output patch decoded through a ResidualBlock to `[n_patches, 128, 10]`; index 0 is point forecast, indices 1–9 are quantiles q0.1–q0.9
