Model
Model explorer

Pixtral Large (124B)

OPEN
Mistral AI · Pixtral family · released Nov 18, 2024

124B multimodal model built on Large 2; strong chart/document/math-vision reasoning; led LMSys vision leaderboard among open-weights models at launch.

ReasoningCodingVisionFunction callingTool useAgentic
1190.3
Elo · unrated
Parameters
124B
Active params
124B (dense)
Context
128K tokens
Architecture
Dense decoder (123B, built on Mistral Large 2) + vision encoder (1B)
License
Mistral Research License (non-commercial; commercial license required for production)
Languages
—
API price (in/out)
No hosted API
Modalities
text · vision
Benchmark results
Bar shows position within the tracked field; marker = field best
AI2DVision93.8%#12
best: Molmo 72B · 96.3%
ChartQAVision88.1%#11
best: MiniMax-VL-01 · 91.7%
DocVQAVision93.3%#18
best: Qwen2-VL-72B · 96.5%
MathVistaVision69.4%#28
best: Seed 2.1 Pro · 90.7%
MMMUVision64.0%#69
best: Claude Fable 5 · 89.3%
VQAv2Vision80.9%#6
best: Reka Edge (2026) · 88.4%
Run it locally
VRAM @ Q4
75 GB
VRAM @ FP16
248 GB
Fits on (Q4)
M3 Max 128GBM3 Ultra 512GBH200 141GBB200 192GB
124B model exceeds a single RTX 4090's 24GB VRAM even at Q4; requires multi-GPU or CPU offload.
Quantizations
GGUF Q4 · MLX
Pixtral family
Elo progression across releases
Pixtral 12BSep 2024617.0Pixtral Large (124B)Nov 20241190.3
Resources
HFmistralai/Pixtral-Large-Instruct-2411DOCSdocs.mistral.ai/models/model-cards/pixtral-large-24-11
API price weights · each benchmark row carries its own source badge (see methodology)