Text Generation
llama-cpp-python
GGUF
llama.cpp
Mixture of Experts
ssd-offload
smallthinker
expert-paging
low-ram
Instructions to use HelloSun/SmallThinker4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use HelloSun/SmallThinker4b with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="HelloSun/SmallThinker4b", filename="{{GGUF_FILE}}", )output = llm( "Once upon a time,", max_tokens=512, echo=True ) print(output)
- Notebooks
- Google Colab
- Kaggle
Download validate/last-run.json from HelloSun/SmallThinker4b: direct link, hf CLI and curl.
- Browser
- Download file 1.53 kB
-
https://huggingface.co/HelloSun/SmallThinker4b/resolve/main/validate/last-run.json
- Command line
-
hf download hf://HelloSun/SmallThinker4b/validate/last-run.json
-
curl -L -o last-run.json https://huggingface.co/HelloSun/SmallThinker4b/resolve/main/validate/last-run.json
1.53 kB
| { | |
| "when": "2026-10-07T16:47:29+0200", | |
| "model": "/root/work/models/SmallThinker-4B-A0.6B-Instruct.Q4_K.gguf", | |
| "port": 8080, | |
| "pid": 38517, | |
| "ram_budget_mb": 512, | |
| "arena_mb": -1875, | |
| "expert_slots": 252, | |
| "chat": { | |
| "prompt_tokens": 37, | |
| "completion_tokens": 16, | |
| "seconds": 3.89, | |
| "tok_per_s": 4.1168, | |
| "content": "A **MoE (Model Parallelism) layer** enables a neural network to", | |
| "timings": { | |
| "cache_n": 0, | |
| "prompt_n": 37, | |
| "prompt_ms": 3122.416, | |
| "prompt_per_token_ms": 84.38962162162163, | |
| "prompt_per_second": 11.849798361268965, | |
| "predicted_n": 16, | |
| "predicted_ms": 744.758, | |
| "predicted_per_token_ms": 49.650533333333335, | |
| "predicted_per_second": 20.140770559027224 | |
| } | |
| }, | |
| "peak": { | |
| "total_rss_gb": 1.0901, | |
| "anon_rss_gb": 0.1931, | |
| "file_rss_gb": 1.0845, | |
| "peak_swap_gb": 0.0, | |
| "hwm_rss_gb": 1.0901 | |
| }, | |
| "io_delta": { | |
| "read_bytes": 1160712192, | |
| "rchar": 0, | |
| "write_bytes": 4096 | |
| }, | |
| "pager": { | |
| "routing_calls": 4608, | |
| "expert_touches": 56128, | |
| "hits": 23134, | |
| "misses": 2146, | |
| "evictions": 2920, | |
| "evicted_bytes": 6264345600, | |
| "sweeps": 537, | |
| "resident_bytes": 535071744, | |
| "budget_bytes": 536870912, | |
| "hit_rate": 0.4122, | |
| "proc_self_read_bytes": 2439696384, | |
| "proc_self_rchar": 11946335 | |
| }, | |
| "pager_expert_resident_mib": 510.3, | |
| "pager_expert_budget_mib": 512.0, | |
| "pager_expert_within_budget": true, | |
| "pager_hit_rate": 0.4122, | |
| "ok": true, | |
| "model_size_mib": 2508 | |
| } | |