Skip to content
Local AI Wizard
Your machine
Theme

Ministral 3 8B vs Llama 3.1 8B for local text

A newer and an older 8B model for an 8 to 16 GB computer

Facts from the reviewed catalog

Facts from the reviewed catalog
SpecificationMinistral 3 8BLlama 3.1 8B Instruct
TaskText & chatText & chat
Parameters8.92B8B
LicenseApache 2.0Publisher terms: Llama 3.1
Commercial useAllowedlicense textAllowed with conditionsOver 700M monthly users needs a Meta license; show "Built with Llama" · license text
Quality · LMArena 2026-10-22 · textNot on the LMArena text leaderboard of 2026-10-22 · source1,211 ± 41,207–1,215 · 49,605 votes · rank 345 · source
MemoryQ4 planning estimate (~7 GB of weights before the machine's reserve)Q4 planning estimate (~6 GB of weights before the machine's reserve)
Checkpointofficial cardofficial card

Only Llama 3.1 8B Instruct has a result in LMArena 2026-10-22 · text (1,211 ± 4); Ministral 3 8B has none, so the two are not ranked from this source and no score is inferred for it. Blind pairwise votes by people on the hosted model, CC BY 4.0. A rating is relative; overlapping intervals mean the order is not settled. Scale: Arena rating (style control, overall) ± half the confidence interval.

Memory on three machines

Each machine's own reserve is added (graphics cards 2 GB; unified and system memory 25%, 2 to 8 GB). Planning estimates, not measured runs.

Ministral 3 8B vs Llama 3.1 8B for local text: memory on three machines
MachineMinistral 3 8BLlama 3.1 8B Instruct
RTX 4060 · 8 GB VRAM~9 GBIn memory, outside the fast part · offload to PC RAM (32 GB assumed)~8 GBFast memory holds it
RTX 3060 · 12 GB VRAM~9 GBFast memory holds it~8 GBFast memory holds it
MacBook Air 13-inch M4 · 16 GB unified~11 GBFast memory holds it~10 GBFast memory holds it

Ministral 3 8B: Q4 planning estimate (~7 GB of weights before the machine's reserve). Llama 3.1 8B Instruct: Q4 planning estimate (~6 GB of weights before the machine's reserve). Formulas

Everything that runs on each machine: RTX 4060 · 8 GB VRAM · RTX 3060 · 12 GB VRAM · MacBook Air 13-inch M4 · 16 GB unified

Which to choose

Ministral 3 8B (December 2025) is about 16 months newer than Llama 3.1 8B (July 2024). It reads images as well as text, publishes a 262,144-token context against 128,000, and is Apache 2.0, while Llama's licence adds conditions for very large services. Both need about 6 to 7 GB of Q4 weights; on an 8 GB card the table puts Ministral just outside the fast memory and Llama inside it. Mistral ships the instruct checkpoint in FP8 and says it fits in 12 GB of VRAM that way. Only Llama has a rating in the LMArena text release; Mistral's card reports its own results.

Files and run commands

The default file of the most downloaded collected source, or the commands each card documents. Not run by this site.

Ministral 3 8B

mistralai/Ministral-3-8B-Instruct-2512-GGUF · Ministral-3-8B-Instruct-2512-Q4_K_M.gguf · Q4_K_M · 5.2 GB · mistralai

ollama run hf.co/mistralai/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
llama-server -hf mistralai/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M
lms get https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512-GGUF
  • mlx-lm: mlx-lm runs MLX repositories; pick the MLX version on a Mac

Llama 3.1 8B Instruct

bartowski/Meta-Llama-3.1-8B-Instruct-GGUF · Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf · Q4_K_M · 4.9 GB · bartowski

ollama run hf.co/bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M
llama-server -hf bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M
lms get https://huggingface.co/bartowski/Meta-Llama-3.1-8B-Instruct-GGUF
  • mlx-lm: mlx-lm runs MLX repositories; pick the MLX version on a Mac

Numbers each publisher reports

Each card's own results with its own settings. Publishers choose different tests and settings, so these values are never compared across publishers here.

Ministral 3 8B · reported by Mistral

14 results · Mistral's model card · read 2026-10-25
AIME25
0.787Reasoning
AIME24
0.860Reasoning
GPQA Diamond
0.668Reasoning
LiveCodeBench
0.616Reasoning
Arena Hard
0.509Instruct
WildBench
66.8Instruct
MATH Maj@1
0.876Instruct
MM MTBench
8.08Instruct
Every reported result14
Multilingual MMLU
0.706Base
MATH CoT 2-Shot
0.626Base · 2-shot
AGIEval 5-shot
0.591Base · 5-shot
MMLU Redux 5-shot
0.793Base · 5-shot
MMLU 5-shot
0.761Base · 5-shot
TriviaQA 5-shot
0.681Base · 5-shot

Values as printed on the card, with the publisher's own settings (card section: Benchmark Results › Reasoning; Benchmark Results › Instruct); column “Ministral 3 8B”. Not comparable across publishers and not used in any ranking. Model card

Llama 3.1 8B Instruct · reported by Meta

24 results · Meta's model card · read 2026-10-25
MMLU
69.4Llama 3.1 8B Instruct · Instruction tuned models · General
MMLU (CoT)
73.0Llama 3.1 8B Instruct · Instruction tuned models · General
MMLU-Pro (CoT)
48.3Llama 3.1 8B Instruct · Instruction tuned models · General
IFEval
80.4Llama 3.1 8B Instruct · Instruction tuned models · General
ARC-C
83.4Llama 3.1 8B Instruct · Instruction tuned models · Reasoning
GPQA
30.4Llama 3.1 8B Instruct · Instruction tuned models · Reasoning
HumanEval
72.6Llama 3.1 8B Instruct · Instruction tuned models · Code
MBPP ++ base version
72.8Llama 3.1 8B Instruct · Instruction tuned models · Code
Every reported result24
Multipl-E HumanEval
50.8Llama 3.1 8B Instruct · Instruction tuned models · Code
Multipl-E MBPP
52.4Llama 3.1 8B Instruct · Instruction tuned models · Code
GSM-8K (CoT)
84.5Llama 3.1 8B Instruct · Instruction tuned models · Math
MATH (CoT)
51.9Llama 3.1 8B Instruct · Instruction tuned models · Math
API-Bank
82.6Llama 3.1 8B Instruct · Instruction tuned models · Tool Use
BFCL
76.1Llama 3.1 8B Instruct · Instruction tuned models · Tool Use
Gorilla Benchmark API Bench
8.2Llama 3.1 8B Instruct · Instruction tuned models · Tool Use
Nexus (0-shot)
38.5Llama 3.1 8B Instruct · Instruction tuned models · Tool Use · 0-shot
Multilingual MGSM (CoT)
68.9Llama 3.1 8B Instruct · Instruction tuned models · Multilingual
MMLU · Portuguese (5-shot, macro_avg/acc)
62.12Llama 3.1 8B · Multilingual benchmarks · General · 5-shot
MMLU · Spanish (5-shot, macro_avg/acc)
62.45Llama 3.1 8B · Multilingual benchmarks · General · 5-shot
MMLU · Italian (5-shot, macro_avg/acc)
61.63Llama 3.1 8B · Multilingual benchmarks · General · 5-shot
MMLU · German (5-shot, macro_avg/acc)
60.59Llama 3.1 8B · Multilingual benchmarks · General · 5-shot
MMLU · French (5-shot, macro_avg/acc)
62.34Llama 3.1 8B · Multilingual benchmarks · General · 5-shot
MMLU · Hindi (5-shot, macro_avg/acc)
50.88Llama 3.1 8B · Multilingual benchmarks · General · 5-shot
MMLU · Thai (5-shot, macro_avg/acc)
50.32Llama 3.1 8B · Multilingual benchmarks · General · 5-shot

Values as printed on the card, with the publisher's own settings (card section: Benchmark scores › Instruction tuned models; Benchmark scores › Instruction tuned models › Multilingual benchmarks). Not comparable across publishers and not used in any ranking. Model card

Limits and dates

The published quality benchmark does not test a local quantized build. Memory estimates are not speed measurements or guaranteed hardware fits. Repository size, when available, is not VRAM. Formulas

Page updated · Catalog reviewed · LMArena 2026-10-22 · text