Mistral Large API Pricing: European Flagship Intelligence and Sovereignty
Comprehensive Mistral Large API pricing analysis ($0.50/M input, $1.50/M output), 128K context economics, European data sovereignty compliance, and multilingual benchmarks.
Full specs, context window and API limits →How much does Mistral Large 3 cost per million tokens?
Mistral Large costs $0.50 per million input tokens and $1.50 per million output tokens ($0.75/M blended at 3:1). Mistral premier model providing frontier reasoning, complex multilingual capabilities, and full European GDPR data residency compliance. Verified 2026-09-08.
How much does Mistral Large 3 cost per 1,000 requests?
Computed from generated token pricing. Each row assumes the listed input and output tokens per request; this model has no measured verbosity factor, so the unadjusted output estimate is shown.
| Request shape | Input tokens | Output tokens | Cost / 1,000 requests |
|---|---|---|---|
| Short | 100 | 50 | $0.1250 |
| Medium | 1,000 | 500 | $1.2500 |
| Long | 4,000 | 2,000 | $5.0000 |
Formula: ((input price × input tokens) + (output price × output tokens × verbosity factor)) ÷ 1,000,000 × 1,000. Assumptions: short 100/50, medium 1,000/500, long 4,000/2,000 input/output tokens per request. Unadjusted — no measured verbosity factor is available.
Three model-specific pricing decisions
Mistral Large owns this family rate audit; no family member is silently substituted when its dated price row is missing.
1. Dated Large / Medium / Small rate audit
| Model | Verified | Input / output per M | Ordering note |
|---|---|---|---|
| Mistral Large 3 | 2026-08-14 | $0.50 / $1.50 | Compare listed rates only |
| Mistral Medium 3 | 2026-08-14 | $1.50 / $7.50 | Compare listed rates only |
| Mistral Small 3.1 | 2026-08-14 | $0.15 / $0.60 | Compare listed rates only |
2. Input/output shape and accepted-result crossover against Medium
Fixed workload boundary
| Fixed workload | Mistral Large 3 | Mistral Medium 3 | Boundary |
|---|---|---|---|
| Coding | $350.00 | $1350.00 | 10% accepted-result uplift needed to justify premium |
| Long document | $520.00 | $1800.00 | 15% accepted-result uplift needed to justify premium |
Formula: calls × (input tokens × input $/M + output tokens × output $/M) ÷ 1,000,000. The uplift threshold is a decision input, not a measured quality claim.
3. Coding cost per passing run
| Prompt / rubric / run date | Model evidence | Cost per passing run |
|---|---|---|
| Median of two sorted arrays · published rubric · 2026-06-21 | Unavailable — model not in eligible dated run | Unavailable |
| Mechanic | Evidence |
|---|---|
| Cache | Unavailable — no Mistral model-specific source |
| Batch | Unavailable — no dated model rule |
| Latency | Unavailable — no eligible dated run |
Verified 2026-08-14. Luna is the data owner for this rendered decision module. “Unavailable” means the current dated registry has no model-specific evidence; it is not a zero. First-party price source · Run this scenario in the playground.
All results are server-rendered for Mistral Large 3; formulas expose fixed inputs and missing evidence remains visibly unavailable.
Batch 62 · exact-model pricing decision contributions · verified 2026-09-07
Exact model boundary: Mistral Mistral Large 3 (mistral-large). Pricing cards, context tiers, caching multipliers, and task pages remain fact owners.
Multilingual enterprise customer support token budgeting
Frozen Batch 62 scenario board. Formula / deterministic rule: support_cost = tickets * ((history_tokens * 0.50 + response_tokens * 1.50) / 1M) Boundary: Owns multilingual enterprise support economics for Mistral Large 3.
| Frozen scenario / field ID | Exact identity and evidence fields | Result | State |
|---|---|---|---|
batch62-mistral-large-m1-r110K multilingual support tickets | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=10K multilingual support tickets; ticket volume; prompt tokens; response tokens; monthly spend; cost per ticket; SLA status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 10K multilingual support tickets is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m1-r250K European enterprise queries | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=50K European enterprise queries; ticket volume; prompt tokens; response tokens; monthly spend; cost per ticket; SLA status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 50K European enterprise queries is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m1-r3200K localized e-commerce chats | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=200K localized e-commerce chats; ticket volume; prompt tokens; response tokens; monthly spend; cost per ticket; SLA status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 200K localized e-commerce chats is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m1-r41M high-volume customer interactions | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=1M high-volume customer interactions; ticket volume; prompt tokens; response tokens; monthly spend; cost per ticket; SLA status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 1M high-volume customer interactions is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m1-r5context overflow past 128K | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=context overflow past 128K; ticket volume; prompt tokens; response tokens; monthly spend; cost per ticket; SLA status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — context overflow past 128K has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
batch62-mistral-large-m1-r6unsupported dialect fallback | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=unsupported dialect fallback; ticket volume; prompt tokens; response tokens; monthly spend; cost per ticket; SLA status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unsupported dialect fallback has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: Mistral AI model pricing; verification date 2026-09-07. Missing or conflicting joins fail closed.
Code generation and automated refactoring cost matrix
Frozen Batch 62 scenario board. Formula / deterministic rule: refactor_cost = files * ((repo_slice_tokens * 0.50 + patch_tokens * 1.50) / 1M) Boundary: Owns software engineering and refactoring economics on Mistral Large.
| Frozen scenario / field ID | Exact identity and evidence fields | Result | State |
|---|---|---|---|
batch62-mistral-large-m2-r1100 function unit test generations | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=100 function unit test generations; engineering tasks; input tokens; completion tokens; total spend; cost per PR; QA verification verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 100 function unit test generations is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m2-r2500 multi-file dependency updates | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=500 multi-file dependency updates; engineering tasks; input tokens; completion tokens; total spend; cost per PR; QA verification verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 500 multi-file dependency updates is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m2-r32K pull request code reviews | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=2K pull request code reviews; engineering tasks; input tokens; completion tokens; total spend; cost per PR; QA verification verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 2K pull request code reviews is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m2-r410K repository static analysis passes | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=10K repository static analysis passes; engineering tasks; input tokens; completion tokens; total spend; cost per PR; QA verification verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 10K repository static analysis passes is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m2-r5syntax check failure re-prompt | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=syntax check failure re-prompt; engineering tasks; input tokens; completion tokens; total spend; cost per PR; QA verification verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — syntax check failure re-prompt has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
batch62-mistral-large-m2-r6unverified code execution sandbox | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=unverified code execution sandbox; engineering tasks; input tokens; completion tokens; total spend; cost per PR; QA verification verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unverified code execution sandbox is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
First-party provenance: Mistral AI documentation; verification date 2026-09-07. Missing or conflicting joins fail closed.
Mistral Large vs Mistral Medium / Small tier allocation
Frozen Batch 62 scenario board. Formula / deterministic rule: portfolio_cost = large_pct * large_cost + med_pct * med_cost + small_pct * small_cost Boundary: Owns multi-tier routing across the Mistral model family.
| Frozen scenario / field ID | Exact identity and evidence fields | Result | State |
|---|---|---|---|
batch62-mistral-large-m3-r1100% Mistral Large 3 baseline | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=100% Mistral Large 3 baseline; tier allocation; monthly queries; blended cost; savings vs pure Large; quality trade-off; recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 100% Mistral Large 3 baseline is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m3-r250% Large / 50% Small 3.1 mix | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=50% Large / 50% Small 3.1 mix; tier allocation; monthly queries; blended cost; savings vs pure Large; quality trade-off; recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 50% Large / 50% Small 3.1 mix is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m3-r320% Large / 80% Small 3.1 mix | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=20% Large / 80% Small 3.1 mix; tier allocation; monthly queries; blended cost; savings vs pure Large; quality trade-off; recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 20% Large / 80% Small 3.1 mix is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m3-r410% Large / 40% Med / 50% Small | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=10% Large / 40% Med / 50% Small; tier allocation; monthly queries; blended cost; savings vs pure Large; quality trade-off; recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 10% Large / 40% Med / 50% Small is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m3-r5edge deployment fallback | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=edge deployment fallback; tier allocation; monthly queries; blended cost; savings vs pure Large; quality trade-off; recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — edge deployment fallback is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch62-mistral-large-m3-r6unresolved model routing rule | model=mistral-large; provider=Mistral; slug=mistral-large; scenario=unresolved model routing rule; tier allocation; monthly queries; blended cost; savings vs pure Large; quality trade-off; recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unresolved model routing rule has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: Mistral AI model pricing; verification date 2026-09-07. Missing or conflicting joins fail closed.
Method and limitations: formulas are deterministic; observed and assumed inputs are labeled; no missing provider, host, account, region, realm, alias, snapshot, revision, weight, artifact, control, tool, modality, workload, rate-period, timestamp, or result is transferred. Run the Mistral Large 3 Batch 62 scenario →
mistral-largeMistral Large API Pricing: European Flagship Intelligence and Sovereignty
Mistral Large costs $0.50 per million input tokens and $1.50 per million output tokens ($0.75/M blended at 3:1). Mistral premier model providing frontier reasoning, complex multilingual capabilities, and full European GDPR data residency compliance. Verified 2026-09-08.
Mistral Large delivers top-tier cognitive performance and multilingual mastery at $0.75/M blended.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Multilingual European contract review (16K in, 3K out): $0.012500 per agreement |
| Scenario 2 | Cross-border EU regulatory compliance audit (24K in, 4K out): $0.018000 per dossier |
| Scenario 3 | French/German/English corporate policy synthesis (8K in, 1.5K out): $0.006250 per policy |
| Scenario 4 | Complex multi-step SQL database optimization (12K in, 2K out): $0.009000 per query set |
| Scenario 5 | Autonomous agent multi-turn planning pass (32K in, 4K out): $0.022000 per planning turn |
| Scenario 6 | Monthly enterprise tier (100M blended tokens): $75.00 infrastructure budget |
Full European data residency ensures regulatory compliance for sensitive enterprise workflows.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Hosted in EU-based data centers under strict European jurisdiction and privacy governance |
| Scenario 2 | Zero training on customer API inputs guarantees enterprise confidentiality and trade secret safety |
| Scenario 3 | Certified GDPR, HIPAA, and ISO/IEC 27001 compliance simplifies enterprise vendor procurement |
| Scenario 4 | Native fluency across French, German, Spanish, Italian, and European regional languages |
| Scenario 5 | Eliminates legal complexities associated with cross-border data transfers to US cloud providers |
| Scenario 6 | Essential compliance foundation for European banking, healthcare, and public sector deployments |
Delivers elite reasoning and multilingual fluency at an 85%+ discount compared to Western flagships.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Mistral Large ($0.75/M blended) vs GPT-5.4 ($5.625/M blended): 86.7% operational cost savings |
| Scenario 2 | Mistral Large vs Claude Sonnet 4.6 ($6.00/M blended): 87.5% operational cost savings |
| Scenario 3 | Competitive performance on MMLU, GSM8K, and HumanEval coding benchmarks |
| Scenario 4 | High-volume production run (1B tokens): saves >$4,800 compared to Western proprietary tiers |
| Scenario 5 | Standard OpenAI-compatible REST API allows seamless drop-in routing replacement |
| Scenario 6 | Delivers the highest cost-to-capability ratio among all non-subsidized frontier LLMs |
How fast is Mistral Large 3?
How much does Mistral Large 3 cost at scale?
| Tokens / month | Est. cost (blended 3:1) |
|---|---|
| 100,000 | $0.07 |
| 1,000,000 | $0.75 |
| 10,000,000 | $7.50 |
| 100,000,000 | $75.00 |
How does Mistral Large 3 compare with other models?
What is Mistral Large 3 best for?
What should you explore next for Mistral Large 3?
Which Mistral Large 3 head-to-head comparisons are available?
What are common questions about Mistral Large 3?
Is Mistral Large 3 cheaper than GPT-5 Mini?
Mistral Large 3 costs $0.75/M blended tokens, GPT-5 Mini costs $0.69/M — GPT-5 Mini is cheaper.
How much does 1 million tokens cost with Mistral Large 3?
At a 3:1 input:output ratio, 1 million blended tokens costs approximately $0.75. Pure input costs $0.50/M; pure output costs $1.50/M.
What does Mistral Large 3 cost at high volume?
At 100 million blended tokens a month, Mistral Large 3 costs approximately $75.00. See the cost-at-scale table below for other volumes.