Mistral AI · Benchmarks

Mistral Large 4 benchmarks

Mistral Large 4 has 2 published scores in this dataset, from a Mistral AI release dated 2026-10-29. It does not yet have scores in enough categories to receive a composite index. Its strongest area is agentic (index 91.7) and its weakest is coding (index 58.1). Its best placing is on Terminal-Bench 4.0, where its 41.0 ranks 9th of 16 models with a published score.

Open weightsOfficial page

Category profile

Index 0–100: each score relative to the best published score on that test.

ReasoningCodingAgenticKnowledge
Reasoning
—no data
Coding
58.11 test · not ranked
Agentic
91.71 test · not ranked
Knowledge
—no data

Model facts

Lab
Mistral AI
Type
Open weights
Released
2026-10-29
Parameters
1T total, 52B active (MoE)
Context
1M
License
Mistral Ecosystem Open-Weight License (modified MIT; MaaS use needs commercial license)

Mistral Large 4 hardware requirements and cost

Even at 4-bit it needs about 690 GB (8× H200 (1,128 GB)), well beyond a home workstation, so most people run it on rented multi-GPU servers. Full BF16 precision needs about 2.3 TB (Multiple nodes (over 1.1 TB)).

Estimated GPU memory for Mistral Large 4
PrecisionWeightsWith headroomSmallest setup
BF16 (full precision)2.0 TB2.3 TBMultiple nodes (over 1.1 TB)
FP8 / 8-bit1.0 TB1.1 TBMultiple nodes (over 1.1 TB)
4-bit (GGUF Q4_K_M)600 GB690 GB8× H200 (1,128 GB)

Estimate: parameters × bytes per parameter, plus 15% for the KV cache and runtime at modest context lengths. Long contexts and large batches need more.

Lighter or stronger open models

Open models ranked just above and below on the composite index, with their 4-bit memory needs.

Mistral Large 4 speed and pricing

Measured figures link to their source; estimates are labelled.

API output speed
112 tok/sOpenRouter median provider, last 30 min · OpenRouter model page provider stats
Time to first token
1.72 sOpenRouter model page provider stats
API price per 1M tokens
$0.700 in · $2.10 outlab's own endpoint on OpenRouter · OpenRouter endpoints API
Compare speed and price across all models

Score by test

Rank is among every model with a published score on that test.

Terminal-Bench 4.041.0%Terminal-Bench 4.0, ML4 OSS release (Preview scored 28)#9 of 16 · 29.6 behind Claude Sonnet 5.5BrowseComp84.5%ML4 OSS release, version not stated (SFT checkpoint 75.0)#5 of 8 · 7.6 behind Kimi K3.1 Pro

Nearby in the composite ranking

Models with the closest composite index.

  1. 1Claude Opus 5.598.4
  2. 2Kimi K3.1 Pro94.4
  3. 3Claude Sonnet 5.591.9
  4. 4GPT-6 Astra91.4
  5. 5Claude Fable 5.191.2
  6. 6DeepSeek V4 Pro91.0
  7. 7Kimi K390.6

Head-to-head comparisons

Side-by-side pages built from the tests both models report.

No head-to-head page for Mistral Large 4 yet. Compare it on the rankings dashboard.

Sources

ModelTestScoreSettingSource
Mistral Large 4Terminal-Bench 4.041.0Terminal-Bench 4.0, ML4 OSS release (Preview scored 28)Mistral AI blog: Mistral Large 4 open weights
Mistral Large 4BrowseComp84.5ML4 OSS release, version not stated (SFT checkpoint 75.0)Mistral AI blog: Mistral Large 4 open weights

Scores are published results from model cards, lab reports, and public leaderboards, each linked to its source. Settings such as thinking mode or test version differ between sources and are shown next to each score. OpenLLM Buddy catalog models are uncensored fine-tunes; their scores are those of the official base model. Data collected 2026-10-30. How the index works