Model reference · open weights

SmolLM3-GSM8K-SFT

SmolLM3-GSM8K-SFT is an open-weight language model from HuggingFaceTB, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.

LLMs HuggingFaceTB 1 variants 317 downloads/mo
Request this model on EU hardware All served models Not on the shared API today — deployed on request.

About

What SmolLM3-GSM8K-SFT is

SmolLM3-3B-GSM8K-SFT Fine-tuned version of HuggingFaceTB/SmolLM3-3B-Base optimized for grade school math (GSM8K benchmark). Performance Quick Start Expected output: Using with vLLM (Recommended for Speed) Training Details Chat Template This model uses the ChatML format: Training History We tried multiple approaches to improve math reasoning: Key finding: More diverse training data (100k vs 50k samples) was more effective than more epochs or GRPO reinforcement learning. Reproduction Training and evaluation scripts are available in the training/ folder: Limitations - Optimized specifically for grade school math; may not generalize to advanced mathematics - Best performance with step-by-step reasoning format ending with #### answer - Context window limited to 2048 tokens during training Citation License Apache 2.0

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

MakerHuggingFaceTB
TypeLanguage models
Parameters (lead)3.1B
Context64k tokens
Variants1
Based onHuggingFaceTB/SmolLM3-3B-Base
Released2026-04-03
Popularity317 downloads / month
Likes2
LicenceOpen weights

How it works

How language models work

Your prompttext / messagesTransformerattention over tokensNext-token loopgenerate + streamResponsetext · tool callsA language model reads your tokens and predicts the next one, again and again, streaming the reply back.

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

VariantParamsPrecisionVRAMFits 16 GBWeights
SmolLM3-3B-GSM8K-SFT3.1BBF16~7.1 GBWeights ↗

Benchmarks

Reported results

As published on the model card — the maker's own numbers, not measured by AxForge.

TaskDatasetMetricScore
Math ReasoningGSM8KGSM8K Accuracy65.8

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys smollm3-gsm8k-sft for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (smollm3-gsm8k-sft below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"smollm3-gsm8k-sft","messages":[{"role":"user","content":"Hello"}]}'

Details

Languages, data & research

Languages

en

Trained / evaluated on

meta-math/MetaMathQA

Tags

safetensors smollm3 math gsm8k sft fine-tuned reasoning text-generation conversational en dataset:meta-math/MetaMathQA model-index deploy:azure

Licence

Open weights

Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗

Sources

Weights & code

Want SmolLM3-GSM8K-SFT on EU-owned hardware?

Request this model on EU hardware See what’s served now

Explore

More language models

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms