Smaug Mini
Abacus AI · abacusai/Smaug-Mini
Q4 file 16.5 GB (Q4_K_M) · runs well from 24 GB · Weights license: Apache 2.0
Quality
Quality · Overall
LiveBench 2026-06-25 · 0–100 · higher is better · this model outlined, #7 of 22
- Color = publisher
- Memory under review
Source: LiveBench 2026-06-25 · collected 2026-09-26 · open models without a score are not drawn (no score ≠ 0)
No result in LMArena 2026-10-22 Not in LMArena 2026-10-22
LiveBench categories
This model's mean per category · 0–100
Source: LiveBench 2026-06-25 · collected 2026-09-26 · open models without a score are not drawn (no score ≠ 0)
Reported by the publisher
- AutomationBench
- 41.8
- JobBench
- 50.5
- LiveBench agentic coding
- 60.8
- NL2Repo-Bench
- 55.8
- GPQA-diamond
- 89.4
- HLE
- 34.2
- IFBench
- 82.0
- LiveBench overall
- 76.9
Every reported result
- MMMU-Pro
- 75.6
Values as printed on the card, with the publisher's own settings (card section: 3. Evaluation › Smaug-Mini vs its base); column “Smaug-Mini”. Not comparable across publishers and not used in any ranking. Model card
Run it
Run it
File Q4_K_M · 16.5 GB · prithivMLmods/Smaug-Mini-GGUF · What is Q4?
ollama run hf.co/prithivMLmods/Smaug-Mini-GGUF:Q4_K_M- Install Ollama
- Run the command above
- Ask:
Explain what a context window is, in two sentences.
Quality × memory
Quality × memory
LiveBench 2026-06-25 overall vs. Q4 planning memory (log scale) · 18 open models
- This model
- Color = publisher
Memory = the model's own figure: parameters × 0.625 bytes + 20% working room, before any machine's reserve; with a chosen machine, the total there with its reserve (25% of memory, 2 to 8 GB; 2 GB per graphics card). An estimate, not measured VRAM. LiveBench 2026-06-25
Runs on
Fast memory holds it 48
The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.
Show all 48
Within memory, outside the fast part 45
Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.
Show all 45
One memory tier short 9
The same machines with the next memory size up would hold it in fast memory.
Show all 9
Smaug Mini is a text model from Abacus AI with 27B parameters. It reads up to 262,144 tokens of context. Its Q4 weights plus working room come to about 21 GB before the machine's reserve; the smallest reference machines whose fast memory holds that are Mac mini M4 · 32 GB unified (~29 GB with its reserve) and RTX 4090 · 24 GB VRAM (~23 GB with its reserve). We collected 3 downloadable GGUF files from one publisher; the most downloaded GGUF source is prithivMLmods's (905 downloads on the Hub). It scores 76.9 overall in LiveBench 2026-06-25.
Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Each card opens its hardware profile. Not a measured fit or speed.
On the reference machines
| Where Q4 sits | ||||
|---|---|---|---|---|
| Mac mini M2 · 8 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~23 GB | Larger than its memory | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| Mac mini M4 · 16 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~25 GB | Larger than its memory | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| Mac mini M4 · 24 GB unifiedLarger than its memoryNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~27 GB | Larger than its memory | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| Mac mini M4 · 32 GB unifiedFast memory holds itQ5_K_M8,192 tokens | ~29 GB | Fast memory holds it | Q5_K_M | 8,192 tokens |
| Mac mini M5 Pro · 64 GB unifiedFast memory holds itQ5_K_M262,144 tokens | ~29 GB | Fast memory holds it | Q5_K_M | 262,144 tokens |
| Mac Studio M5 Max · 128 GB unifiedFast memory holds itQ5_K_M262,144 tokens | ~29 GB | Fast memory holds it | Q5_K_M | 262,144 tokens |
| Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unifiedFast memory holds itQ5_K_M262,144 tokens | ~29 GB | Fast memory holds it | Q5_K_M | 262,144 tokens |
| RTX 4060 · 8 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~23 GB | In memory, outside the fast part | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| RTX 3060 · 12 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~23 GB | In memory, outside the fast part | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| RTX 5080 · 16 GB VRAMIn memory, outside the fast partNo file on the site's quant ladder fits the fast memory at 8,192 tokensNo file in the previous column | ~23 GB | In memory, outside the fast part | No file on the site's quant ladder fits the fast memory at 8,192 tokens | No file in the previous column |
| RTX 4090 · 24 GB VRAMFast memory holds itQ4_K_M16,384 tokens | ~23 GB | Fast memory holds it | Q4_K_M | 16,384 tokens |
| RTX 5090 · 32 GB VRAMFast memory holds itQ5_K_M65,536 tokens | ~23 GB | Fast memory holds it | Q5_K_M | 65,536 tokens |
| DGX H100 · 8 × 80 GBFast memory holds itQ5_K_M262,144 tokens | ~37 GB | Fast memory holds it | Q5_K_M | 262,144 tokens |
The file column uses the site's quant ladder (Q8_0 to Q3_K_M, MLX 8 to 4-bit on a Mac): the largest collected file whose size × 1.2 plus the context cache and the machine's reserve fits the fast memory. Planning estimates, not measured runs.
Consider this agent-oriented fine-tune of Qwen3.8-27B when you need to compare tool use with the base model.
The publisher documents vLLM serving and agent-specific settings, but this site has not verified a hardware minimum or reproduced its tests.
Versions to download
Files you can download for this checkpoint. The section opens on the most downloaded collected GGUF source, prithivMLmods (905 Hub downloads), at Q4_K_M. Sizes are the Hub's published bytes. What is Q4?
GGUF · the only format collected
prithivMLmods · 905 downloads
Community upload outside the publisher allowlist: No allowlisted publisher ships a GGUF or MLX build of abacusai/Smaug-Mini (Hub search base_model:quantized:abacusai/Smaug-Mini, 2026-10-17); this is the only GGUF with named quants and its README front matter declares base_model: abacusai/Smaug-Mini.
prithivMLmods/Smaug-Mini-GGUF · Smaug-Mini.Q4_K_M.gguf · 16.5 GB
Run this file
ollama run hf.co/prithivMLmods/Smaug-Mini-GGUF:Q4_K_Mllama-server -hf prithivMLmods/Smaug-Mini-GGUF:Q4_K_Mlms get https://huggingface.co/prithivMLmods/Smaug-Mini-GGUF- mlx-lm: mlx-lm runs MLX repositories; pick the MLX version on a Mac
Quality scores on this page belong to the original checkpoint; these files are not scored or ranked. File sizes read 2026-10-23 · download counts read 2026-10-17 · commands: Hugging Face Hub docs: Use Ollama with any GGUF model · llama.cpp server README (-hf <user>/<model>[:quant]) · LM Studio docs: lms get with a Hugging Face URL · mlx-lm README (mlx_lm.generate --model)
Fine-tunes, adapters and other versions
- Quantized versions
- 6
- Fine-tunes
- 0
- Adapters · merges
- 0 · 1
Browse them on the Hugging Face model tree · read 2026-10-23
Uncensored and abliterated versions
Some community uploads remove the model's safety refusals. Quality may drop, the original publisher does not support them, claims about them are the uploader's, and the original license still applies. This site never recommends or ranks them.
0 derivative repositories have “uncensored” or “abliterated” in their name (read 2026-10-23).
Measured speed
No published measurement for this checkpoint
Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.
How to run
~21 GB of Q4 weights plus working room, before the machine's reserve. Formula: parameters × 0.625 bytes, plus 20% working allowance; each machine then adds its system reserve: 25% of unified or system memory, 2 to 8 GB, or 2 GB per GPU on dedicated cards. This is neither measured VRAM nor checkpoint download size.
Q4 planning estimate only. Check checkpoint format, runtime support, KV cache, context, batch size and offload before choosing hardware.
Publisher's example
vllm serve abacusai/Smaug-Mini --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coderEvidence level: Generic estimate only; no execution reproduced by this site.
- Publisher sampling defaults
- temperature 1 · top_p 0.95 · top_k 20generation_config.json · read 2026-10-23source
- Chat template markers
- tool calls · thinkingmarkers found in the template, not a test of the feature · read 2026-10-23source
Before downloading
- Check the exact checkpoint and license linked here.
- Confirm a compatible runtime and precision or quantization for your platform.
- Allow for context, batch size and any extra media components beyond the weights.
Technical details
What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.
- Checkpoint / version
- abacusai/Smaug-Mini
- Task and modality
- language and reasoning · text + image
- Count and basis
- 27B · manufacturer specification · 27B active
- Context
- 262,144 tokens published for this model
- Architecture
- Dense
- Runtime and precision
- Publisher documents vLLM. Precision and local execution are not verified by this site.
- Commercial use
- Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
Every stored field
- Revision checked
- ee82c1ded4be75e160c0f7730fdae9e93ee4c80e
- Architecture
- Qwen3_5ForConditionalGeneration
- Library
- transformers
- Hub task tag
- image-text-to-text
- Formats
- safetensors
- Repository created
- 2026-09-02creation is not the model's release
- Repository updated
- 2026-09-09a commit is not a release date
- Hub downloads (30 days) · likes
- 817 · 13this repository only · read 2026-10-25
- Read from the Hub page
- task, transformers class, task tags, file formatsHub overview and file listing · read 2026-10-18
- Catalog state
- Hub inventory current
This is an agent-oriented fine-tune of Qwen3.8-27B. Its official card gives vLLM tool-use settings, but a benchmark score is not a measurement of tool reliability on your own tasks.
Sources
Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.
- Publisher checkpoint and model card
- Weight license and use terms
- Quality benchmark source · LiveBench 2026-06-25 overall
- Publisher runtime example
Catalog reviewed: 2026-09-30. Benchmark release: 2026-06-25; snapshot collected: 2026-09-26.



























































