LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Mistral Large 4

mistral·mistral/mistral-large-4·BETA·Closed·mistral-large series·NEW
Mistral's 1T-parameter multimodal MoE for reasoning, coding, and agentic work

Specs & pricing

Input / output per 1M tokens
Official price·Mistral
$0.70 / $2.10
Blended $1.05 · Cache read $0.07
Lowest paid·Standard ComputeGateway
$0.68 / $2.09
Blended $1.03 · 2.3× spread
Context
1,048,576
Output limit
262,144
Knowledge cutoff
—
Released / updated
2026-10-06 / 2026-10-06
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF

Quality & performanceArtificial Analysis · Intelligence Index v5 · rep. tier High

Intelligence26.1
Value25
Engineering33.4
Finance & Accounting39.1
Legal34.6
Healthcare & Medical36.7
Strategy & Ops43.3
Economics35.6
Output speed73 tok/s
TTFT1.44 s
Cost per task$2.06
Value formulaIQ 26.1 ÷ min blended $1.032 = 25
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
MediumIQ 23.1 · 67 tok/s
HighIQ 26.1 · 73 tok/s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 19 providers19 with public prices

  • ProviderInput / Output
  • Standard ComputeGateway
    mistralai/mistral-large-4-0
    $0.68 / $2.09
    Cache read $0.07Context 1,048,576Output limit 50,000
  • Ollama CloudCloud
    $0.68 / $2.09
    Cache read $0.07Context 1,048,576Output limit 262,144
  • RequestyGateway
    mistral-large-4@eu
    $0.68 / $2.09
    Cache read $0.07Context 524,288 ⚠Output limit 262,144
  • RequestyGateway
    $0.68 / $2.09
    Cache read $0.07Context 524,288 ⚠Output limit 262,144
  • OpenCodeGateway
    $0.68 / $2.09
    Cache read $0.07Context 524,288 ⚠Output limit 262,144
  • NanoGPTGateway
    $0.68 / $2.09
    Cache read $0.07Context 524,288 ⚠Output limit 262,144
  • MistralOfficialFirst-party
    mistral-large-4-0-8
    $0.70 / $2.10
    Cache read $0.07Context 1,048,576Output limit 262,144beta
  • MistralOfficialFirst-party
    mistral-large-latest
    $0.70 / $2.10
    Cache read $0.07Context 1,048,576Output limit 262,144beta
  • MistralOfficialFirst-party
    $0.70 / $2.10
    Cache read $0.07Context 1,048,576Output limit 262,144beta
  • Vercel AI GatewayCloud
    $0.70 / $2.10
    Cache read $0.07Context 1,048,576Output limit 262,144
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Standard Compute
mistralai/mistral-large-4-0
Gateway$0.68$2.09$0.07—1,048,57650,000
Ollama CloudCloud$0.68$2.09$0.07—1,048,576262,144
Requesty
mistral-large-4@eu
Gateway$0.68$2.09$0.07—524,288 ⚠262,144
RequestyGateway$0.68$2.09$0.07—524,288 ⚠262,144
OpenCodeGateway$0.68$2.09$0.07—524,288 ⚠262,144
NanoGPTGateway$0.68$2.09$0.07—524,288 ⚠262,144
MistralOfficial
mistral-large-4-0-8
First-party$0.70$2.10$0.07—1,048,576262,144beta
MistralOfficial
mistral-large-latest
First-party$0.70$2.10$0.07—1,048,576262,144beta
MistralOfficialFirst-party$0.70$2.10$0.07—1,048,576262,144beta
Vercel AI GatewayCloud$0.70$2.10$0.07—1,048,576262,144

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

effort = noneeffort = higheffort = loweffort = mediumeffort = max

Interleaved thinking (reasoning between tool calls) is declared by 2 of 19 providers.

1 / 19 providers expose no reasoning control (reasoning_options: []).

Your usage cost

Common scenarios
1Standard Compute$167.30
2Ollama Cloud$167.30
3Requesty$167.30
4Requesty$167.30
5OpenCode$167.30
6NanoGPT$167.30
7Mistral · Official$169.40
Switch to Standard Compute to save $2.10/mo (1%). The gap is small, so staying on the official channel is fine.
Note: this is a gateway; verify availability and rate limits yourself.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$4.18$02026-10-072026-10-30
2026-10-30Input list $0.70Output list $2.10Min blended $1.03

Artificial Analysis evaluations15 items

HLE-Diamond17.3%
CritPt10.0%
CritPt v2 Hard27.0%
FrontierCode 2.015.0%
Terminal-Bench 4.024.8%
Terminal-Bench-Science4.3%
AutomationBench-AA57.8%
AA-Briefcase1348
GDP.pdf17.6%
MLCR-AA17.8%
AA-Omniscience Index5.1
AA-Omniscience Accuracy26.0%
AA-Omniscience Non-Hallucination71.8%
mka0.3506
vials-aa0.205

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and scale (percent correct, points or an index), so rows are not comparable with one another.

Related models

Mistral Large 3same series$0.50 / $1.50Mistral Large (latest)same series—Mistral Large 2.1same series—GLM-5.3-Flashcheaper alternative$0.15 / $0.50DeepSeek V4.1 Flash 1017cheaper alternative$0.15 / $0.60GPT-5.6 Lunacheaper alternative$0.20 / $1.20MiniMax-M3cheaper alternative$0.30 / $1.20
Data partly from models.dev (MIT) · Mistral official docs ↗