LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Mistral Large 3
◆ OPEN
by Mistral
OPEN WEIGHTSLOCALAPICLOSEDLONG CONTEXTPAID
RANKED #124 OF 232 ASSISTANTS · OVERALL #201 OF 6,578 · VIBE SCORE 5.8 · 68 VOICES
Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.
HAPPY9.2
16 mentions
SAD5.8
7 mentions
VIBE TRAJECTORY
26 weeks · 55 voices
HOLDING UP
Recent vibe is steady. No clear fade in the crowd.
HOW TO USE IT
Open weights. You can use a hosted service, or download it and run it yourself, free.
RUN IT ON YOUR OWN MACHINE
Open weights: download Mistral Large 3 once and it's yours. No subscription, no rate limits, works offline.
675B total parameters (sparse MoE with 41B active), a Q4 file is about 380 GB to 400 GB
- BARE MINIMUM
Custom multi-GPU workstation with 4x Nvidia GeForce RTX 3090
96 GB VRAM AND 512 GB SYSTEM RAM
~$3800 USEDToo big for standard consumer hardware, requires multi-GPU or hybrid RAM offloading for IQ2_XS or heavy CPU offload at ~
- SWEET SPOT
Dual Apple Mac Studio M2 Ultra cluster or high-end workstati
192 GB VRAM AND 512 GB SYSTEM RAM
~$14500 NEWQ4 quantization with partial offloading, ~12 tok/s, 32k context
- FULL POWER
8x Nvidia H100 80GB SXM server node
640 GB VRAM
~$280000 NEWNVFP4 or FP8 native precision, ~45 tok/s, 256k full context
LEARN HOW TO SET IT UP
WILL MY BOX RUN MISTRAL LARGE 3?
The three cards above are researched picks. Search the machine you actually have — we only say yes if it should reply at a usable speed, not just load the weights and crawl.
HARDWARE PICKS & PRICES ARE RESEARCHED FROM THE LIVE WEB AND REFRESHED AUTOMATICALLY. TREAT THEM AS BALLPARK, NOT GOSPEL.
WHAT CAN YOU ACTUALLY DO WITH IT?
Everyday jobs on each of those rigs: how long Mistral Large 3 takes, and how good the crowd says it is at that kind of work.
| THE JOB | BARE MINIMUMCustom multi-GPU workstation with 4x Nvidia GeForce RTX 3090 | SWEET SPOTDual Apple Mac Studio M2 Ultra cluster or high-end workstati | FULL POWER8x Nvidia H100 80GB SXM server node |
|---|---|---|---|
| Write an emaila paragraph or two, drafted from a one-line brief | ~4.2 MIN★★★★★ | ~42 S★★★★★ | ~11 S★★★★★ |
| Summarize a documenta long report boiled down to the points that matter | ~13 MIN★★★★★ | ~2.1 MIN★★★★★ | ~33 S★★★★★ |
| Build a websitea small landing page, markup and styles together | ~67 MIN★★★★★ | ~11 MIN★★★★★ | ~3 MIN★★★★★ |
| Build a backendan API with routes, storage and tests | ~3.5 HR★★★★★ | ~35 MIN★★★★★ | ~9.3 MIN★★★★★ |
| Build a gamea playable browser game in one file | ~2.1 HR★★★★★ | ~21 MIN★★★★★ | ~5.6 MIN★★★★★ |
STARS ARE THE COMMUNITY SCORE FOR THAT KIND OF WORK, DOCKED FOR HOW SQUASHED THE WEIGHTS GET AT EACH TIER. TIMES ARE COMPUTED FROM RESEARCHED SPEEDS. BALLPARK, NOT GOSPEL.
WHAT THE INTERNET SAYS
Not enough discussion about Mistral Large 3 yet to call it. We found 68 posts but people did not say much either way.
62 POSTS · 6 COMMENTS · LEMMY · REDDIT · GITHUB · BLUESKY · DEV.TO · X · BLOG
PEOPLE LOVE
- No fan mail yet. The crowd is still warming up.
PEOPLE COMPLAIN ABOUT
- No complaints yet. The internet is still looking for one. Suspicious.
LOOKING FOR AN ALTERNATIVE?
Other models the crowd has fully reviewed, starting with multimodal models like this one.
Grok 4.6
Qwen3.8 27B
Gemini 3.7 Flash> SOURCES (12)
- REDDIT · 2025-12-02 · 583 UPVOTESIntroducing Mistral 3
Introducing Mistral 3
- REDDIT · 2026-03-07 · 295 UPVOTESMistral Large 3 performs better that GPT5.3 for OpenClaw
Mistral Large 3 performs better that GPT5.3 for OpenClaw
- REDDIT · 2025-12-02 · 291 UPVOTESIntroducing Mistral 3
Introducing Mistral 3
- REDDIT · 2025-12-01 · 212 UPVOTESMistral seems to be about to release 2 new models: Ministral 3 and Mistral Large 3
Mistral seems to be about to release 2 new models: Ministral 3 and Mistral Large 3
- REDDIT · 2026-04-29 · 137 UPVOTESWow, LeChat in Work mode is actually good and usable!
Wow, LeChat in Work mode is actually good and usable!
- REDDIT · 2026-04-11 · 118 UPVOTESBeen running a fully European AI stack on OpenClaw and honestly it's underrated
Been running a fully European AI stack on OpenClaw and honestly it's underrated
- REDDIT · 2025-12-02 · 117 UPVOTESMistral Large 3 available on AWS Bedrock!
Mistral Large 3 available on AWS Bedrock!
- REDDIT · 2026-03-21 · 111 UPVOTESFree Model List (API Keys)
Free Model List (API Keys)
- REDDIT · 2025-12-02 · 95 UPVOTESIt's been a big week for AI ; Here are 10 massive changes you might've missed:
It's been a big week for AI ; Here are 10 massive changes you might've missed:
- REDDIT · 2026-01-07 · 89 UPVOTESGPT-5.2 is the new champion of the Elimination Game benchmark, which tests social reasoning, strategy, and deception in a multi-LLM…
GPT-5.2 is the new champion of the Elimination Game benchmark, which tests social reasoning, strategy, and deception in a multi-LLM environment. Claude Opus 4.5 and Gemini 3 Flash Preview also made very strong debuts.
- REDDIT · 2026-03-10 · 89 UPVOTESLLM Sycophancy Benchmark: Opposite-Narrator Contradictions. Same dispute, opposite first-person perspectives. Does the model keep the same…
LLM Sycophancy Benchmark: Opposite-Narrator Contradictions. Same dispute, opposite first-person perspectives. Does the model keep the same judgment or start agreeing with whoever is speaking?
- REDDIT · 2025-12-02 · 76 UPVOTESMistral 3 Family Released (10 Models): Large 3 hits 1418 Elo, Ministral 3 (3B/8B/14B) beats Qwen-VL. Full Benchmarks & Specs.
Mistral 3 Family Released (10 Models): Large 3 hits 1418 Elo, Ministral 3 (3B/8B/14B) beats Qwen-VL. Full Benchmarks & Specs.