Skip to content
localllm
日本語

RecursiveSI

flint

Parameters
39.8B
Max context length
262,144
Architecture
flint
License
Commercial OK
Base model
RecursiveSI/flint
License
apache-2.0
llama.cpp
b11803+

About flint

flint is a GGUF model based on RecursiveSI/flint on Hugging Face. GGUF files are published by 3 sources (unsloth/flint-GGUF, lmstudio-community/flint-GGUF, ggml-org/Flint-GGUF). It has about 39.8B parameters, the flint architecture, a maximum context length of 262,144 tokens (values written in the GGUF).

There are 12 quantizations, with file sizes from 11.4 GB (UD-IQ1_M) to 79.6 GB (BF16) (actual sizes on Hugging Face).

Running UD-IQ1_M (11.4 GB) with an 8,192-token context needs an estimated 12.8 GB of memory, and an estimated 13.5 GB at 32,768 tokens. For 8,192 tokens, a GPU with 12 GB or more of VRAM (such as GeForce RTX 3080 Ti, GeForce RTX 5070), or a Mac with 12 GB or more of unified memory, is a rough guide (macOS caps how much memory the GPU can use, so leave some headroom).

The original model is licensed under Apache License 2.0, which allows commercial use (keep the license text and copyright notice when redistributing). Check the original license for the terms of use.

File size and estimated memory per quantization

  • Up to 8 GB
  • Up to 16 GB
  • Up to 24 GB
  • Up to 48 GB
  • Up to 128 GB
  • More

File sizes are the actual values on Hugging Face. Memory is estimated for each context length (KV cache f16). The formula is in How memory is calculated.

All memory figures are Estimated (KV cache f16). It should run on a setup with at least this much VRAM (or Mac unified memory). Click a value to see the devices and models that fit that amount (Find by GPU / Mac).

3 sources publish GGUF files of this model. Quantization types and file sizes differ slightly between sources, so there is a table for each source below.

About the "quality guide" (how far a quantization drifts from the original)

For each quantization type, how far the output drifts from the original model, grouped into six levels from values llama.cpp measured on Llama-3-8B.

The value is mean KLD (the difference between the original and quantized model in next-token probabilities); smaller means closer to the original. Quantizations made with an imatrix (importance data) come out slightly smaller. Levels use the value without an imatrix (or with one, for types that only have that).

These values were not measured on this model. Larger models and MoE models tend to drift less with the same type. Distributor-specific mixed types (such as UD-Q4_K_XL) cannot be matched by type name alone, so they show "-".

Quantization Quality guide KLD (no imatrix) KLD (imatrix) Size (Llama-3-8B)
Q8_0 Nearly identical 0.0014 - 7.96 GiB
Q6_K Very small 0.0055 - 6.14 GiB
Q5_K_M Very small 0.0108 - 5.33 GiB
Q5_K_S Small 0.0166 - 5.21 GiB
Q5_1 Small 0.0180 - 5.65 GiB
Q5_0 Small 0.0222 - 5.21 GiB
Q4_K_M Small 0.0313 0.0282 4.58 GiB
Q4_K_S Medium 0.0431 0.0320 4.37 GiB
IQ4_NL Small - 0.0357 4.35 GiB
IQ4_XS Small - 0.0363 4.14 GiB
Q4_1 Medium 0.0717 - 4.78 GiB
Q4_0 Medium 0.0719 - 4.34 GiB
Q3_K_L Medium 0.0872 0.0731 4.03 GiB
Q3_K_M Medium 0.1019 0.0844 3.74 GiB
IQ3_M Medium - 0.1025 3.53 GiB
IQ3_S Medium - 0.1113 3.42 GiB
IQ3_XS Large - 0.1387 3.28 GiB
IQ3_XXS Large - 0.1836 3.05 GiB
Q3_K_S Large 0.2282 0.1998 3.41 GiB
IQ2_M Large - 0.3260 2.74 GiB
Q2_K Very large 0.4451 0.3322 2.96 GiB
Q2_K_S Very large - 0.4034 2.96 GiB
IQ2_S Very large - 0.4392 2.56 GiB
IQ2_XS Very large - 0.5463 2.43 GiB
IQ2_XXS Very large - 0.8120 2.24 GiB
IQ1_M Very large - 1.3931 2.01 GiB
IQ1_S Very large - 2.2546 1.88 GiB

Source: llama.cpp (MIT), measured on Llama-3-8B against FP16, Wikitext-2, revision f364eb6f. Original table · Reading and choosing quantizations

GGUF source

unsloth/flint-GGUF

Converted and published by a third party About this label

Last checked: Nov 11, 2026

Architecture
flint
llama.cpp b11803 and later
Parameters
39.8B
Max context length
262,144
License
apache-2.0 Commercial OK
File size and memory per quantization for flint (unsloth/flint-GGUF)
Quantization File size Measured Quality guide 4,096 8,192 16,384 32,768 65,536 131,072 262,144 max
UD-IQ1_M 11.4 GB - 12.6 GB 12.8 GB 13.0 GB 13.5 GB 14.4 GB 16.3 GB 20.0 GB
UD-IQ2_XXS 11.9 GB - 13.1 GB 13.3 GB 13.5 GB 14.0 GB 14.9 GB 16.8 GB 20.5 GB
UD-Q2_K_XL 13.6 GB - 14.8 GB 15.0 GB 15.2 GB 15.7 GB 16.6 GB 18.5 GB 22.2 GB
UD-IQ3_XXS 14.5 GB - 15.8 GB 15.9 GB 16.2 GB 16.6 GB 17.6 GB 19.5 GB 23.2 GB
UD-Q3_K_XL 18.5 GB - 19.8 GB 19.9 GB 20.1 GB 20.6 GB 21.5 GB 23.4 GB 27.2 GB
UD-Q4_K_XL 25.1 GB - 26.3 GB 26.5 GB 26.7 GB 27.2 GB 28.1 GB 30.0 GB 33.7 GB
UD-Q5_K_XL 29.9 GB - 31.2 GB 31.3 GB 31.6 GB 32.0 GB 33.0 GB 34.8 GB 38.6 GB
UD-Q6_K_XL 36.3 GB - 37.6 GB 37.7 GB 38.0 GB 38.4 GB 39.4 GB 41.2 GB 45.0 GB
Q8_0 42.3 GB Nearly identical 43.6 GB 43.7 GB 43.9 GB 44.4 GB 45.3 GB 47.2 GB 51.0 GB
BF16 (3 parts) 79.6 GB Original precision 80.9 GB 81.0 GB 81.2 GB 81.7 GB 82.6 GB 84.5 GB 88.3 GB 
Ollama  ollama run hf.co/unsloth/flint-GGUF:UD-IQ1_M
llama.cpp  llama-server -hf unsloth/flint-GGUF:UD-IQ1_M

GGUF source

lmstudio-community/flint-GGUF

Converted and published by a third party About this label

Last checked: Nov 11, 2026

Architecture
flint
llama.cpp b11803 and later
Parameters
39.8B
Max context length
262,144
License
apache-2.0 Commercial OK
File size and memory per quantization for flint (lmstudio-community/flint-GGUF)
Quantization File size Measured Quality guide 4,096 8,192 16,384 32,768 65,536 131,072 262,144 max
Q4_K_M 24.2 GB Small 25.4 GB 25.6 GB 25.8 GB 26.3 GB 27.2 GB 29.1 GB 32.8 GB
Q6_K 32.7 GB Very small 34.0 GB 34.1 GB 34.4 GB 34.8 GB 35.8 GB 37.6 GB 41.4 GB
Q8_0 42.4 GB Nearly identical 43.6 GB 43.8 GB 44.0 GB 44.5 GB 45.4 GB 47.3 GB 51.0 GB 
Ollama  ollama run hf.co/lmstudio-community/flint-GGUF:Q4_K_M
llama.cpp  llama-server -hf lmstudio-community/flint-GGUF:Q4_K_M

GGUF source

ggml-org/Flint-GGUF

Converted by the llama.cpp developers About this label

Last checked: Nov 11, 2026

Architecture
flint
llama.cpp b11803 and later
Parameters
39.8B
Max context length
262,144
License
apache-2.0 Commercial OK
File size and memory per quantization for flint (ggml-org/Flint-GGUF)
Quantization File size Measured Quality guide 4,096 8,192 16,384 32,768 65,536 131,072 262,144 max
Q4_K_M 23.4 GB Small 24.7 GB 24.8 GB 25.0 GB 25.5 GB 26.4 GB 28.3 GB 32.1 GB
Q8_0 42.4 GB Nearly identical 43.6 GB 43.8 GB 44.0 GB 44.5 GB 45.4 GB 47.3 GB 51.0 GB
BF16 79.6 GB Original precision 80.9 GB 81.0 GB 81.3 GB 81.7 GB 82.7 GB 84.5 GB 88.3 GB 
Ollama  ollama run hf.co/ggml-org/Flint-GGUF:Q4_K_M
llama.cpp  llama-server -hf ggml-org/Flint-GGUF:Q4_K_M

License and commercial use

Apache License 2.0 Commercial OK

Allows commercial use, modification and redistribution. When redistributing, keep the license text and copyright notice, and note changes in modified files.

Based on the license on the original model's Hugging Face page. Original license text (checked Oct 18, 2026)

The summary states facts this site read from the original text and is not legal advice. Always check the original license and its latest terms before use. All licenses

How to run

Mainline llama.cpp can load this model's format (flint) from b11803 (Oct 21, 2026) onward. Ollama and LM Studio bundle llama.cpp internally, so an older app fails to load it with errors such as "unknown model architecture". If that happens, update the app.

Example commands using UD-IQ1_M. To use another quantization, replace the name with one from the table above. Commands for files from other sources are in each source's section. How to run local LLMs

Ollama

Run it by pointing directly at the GGUF on Hugging Face.

ollama run hf.co/unsloth/flint-GGUF:UD-IQ1_M

llama.cpp

Pass the repository and quantization name to llama-server (or llama-cli) with -hf.

llama-server -hf unsloth/flint-GGUF:UD-IQ1_M

LM Studio

Paste unsloth/flint-GGUF into the search box, download it, and choose the quantization from the file list.

Frequently asked questions

How much memory do I need to run flint (UD-IQ1_M)?
The file size is 11.4 GB (measured). With a 8,192-token context, the estimated memory including the KV cache is 12.8 GB. The formula is published at https://localllm.az-arte.com/en/guide/memory-calculation.
Can flint be used commercially?
The original model's license is Apache License 2.0 (Commercial use allowed).Allows commercial use, modification and redistribution. When redistributing, keep the license text and copyright notice, and note changes in modified files. Check the original license text before deciding.
Does flint run in 8 GB, 12 GB, 16 GB or 24 GB of VRAM?
8 GB: no quantization fits; 12 GB: up to UD-IQ1_M (est. 12.8 GB); 16 GB: up to UD-IQ3_XXS (est. 15.9 GB); 24 GB: up to UD-Q3_K_XL (est. 19.9 GB) (estimated at 8,192 tokens with an f16 KV cache, files from the most-downloaded source).
What is the command to run flint with Ollama?
Run "ollama run hf.co/unsloth/flint-GGUF:UD-IQ1_M" to use the GGUF on Hugging Face directly. To use another quantization, replace the name at the end.