LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Mistral Small 4

mistral·mistral/mistral-small-2603·GA·Open weights·mistral-small series
Fast Mistral production model for chat, extraction, and cost-sensitive agents

Specs & pricing

Input / output per 1M tokens
Official price·Mistral
$0.15 / $0.60
Blended $0.26 · Cache read $0.015
Lowest paid·MistralFirst-party
$0.15 / $0.60
Blended $0.26 · 3.9× spread
Context
262,144
Output limit
256,000
Knowledge cutoff
2025-06
Released / updated
2026-03-16 / 2026-03-16
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF
Weights
Hugging Face

Mistral Small 4 is offered by 21 providers on this page. Its official list price is $0.15 per 1M input tokens and $0.60 per 1M output tokens. No paid channel is priced below the official list price. Prices vary about 3.9× across the providers that host it.

Artificial Analysis rates it 4.3 on the Intelligence Index, with 13.3 for Engineering, its strongest domain. Against its lowest blended price of $0.263 per 1M, that is roughly 16 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 146 tokens per second, with 0.87s to the first token. Running one task of the Artificial Analysis suite costs about $0.0399, which reflects how many tokens its reasoning consumes rather than the unit price alone.

The context window is 262,144 tokens at the reference host, but hosts report different limits, from 128,000 to 262,144, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and PDF. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2025-06.

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

Intelligence4.3
Value16
Engineering13.3
Finance & Accounting11.6
Legal11.6
Healthcare & Medical8.8
Strategy & Ops6.8
Economics11.7
Output speed146 tok/s
TTFT0.87 s
Cost per task$0.04
Value formulaIQ 4.3 ÷ min blended $0.263 = 16

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

Available at 21 providers21 with public prices

  • ProviderInput / Output
  • MistralOfficialFirst-party
    $0.15 / $0.60
    Cache read $0.015Context 262,144Output limit 256,000
  • LLM GatewayGateway
    $0.15 / $0.60
    Context 262,144Output limit 256,000
  • DevPass (LLM Gateway)Gateway
    $0.15 / $0.60
    Context 262,144Output limit 256,000
  • OpenRouterGateway
    $0.15 / $0.60
    Cache read $0.015Context 262,144Output limit 209,715
  • SAP AI CoreCloud
    mistralai--mistral-small
    $0.15 / $0.60
    Context 128,000 ⚠Output limit 128,000deprecated
  • Eden AIGateway
    $0.15 / $0.60
    Cache read $0.015Context 262,144Output limit 256,000
  • Kilo GatewayGateway
    $0.15 / $0.60
    Cache read $0.015Context 262,144Output limit 209,715
  • EmpirioLabs AIGateway
    mistral-small-4
    $0.15 / $0.60
    Cache read $0.15Context 262,144Output limit 65,536
  • CortecsGateway
    $0.16 / $0.63
    Cache read $0.016Context 262,144Output limit 256,000
  • LowRouterGateway
    $0.16 / $0.63
    Cache read $0.016Context 256,000 ⚠Output limit 256,000
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
MistralOfficialFirst-party$0.15$0.60$0.015—262,144256,000
LLM GatewayGateway$0.15$0.60——262,144256,000
DevPass (LLM Gateway)Gateway$0.15$0.60——262,144256,000
OpenRouterGateway$0.15$0.60$0.015—262,144209,715
SAP AI Core
mistralai--mistral-small
Cloud$0.15$0.60——128,000 ⚠128,000deprecated
Eden AIGateway$0.15$0.60$0.015—262,144256,000
Kilo GatewayGateway$0.15$0.60$0.015—262,144209,715
EmpirioLabs AI
mistral-small-4
Gateway$0.15$0.60$0.15—262,14465,536
CortecsGateway$0.16$0.63$0.016—262,144256,000
LowRouterGateway$0.16$0.63$0.016—256,000 ⚠256,000

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 = maxToggle (on / off)

Interleaved thinking (reasoning between tool calls) is declared by 2 of 21 providers.

5 / 21 providers expose no reasoning control (reasoning_options: []).

Your usage cost

Common scenarios
1Mistral · Official$43.80
2OpenRouter$43.80
3Eden AI$43.80
4Kilo Gateway$43.80
5Cortecs$45.65
6LowRouter$45.81
The cheapest paid channel is official.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.60$02026-08-052026-10-22
2026-10-22Input list $0.15Output list $0.60Min blended $0.26

Benchmark3 items

NameConditionsScoreMetricSource
Artificial Analysis Coding Index—24.3indexSource ↗
SciCode—38percent correctSource ↗
Terminal-Bench Hard—17.4success rateSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Artificial Analysis evaluations25 items

GPQA Diamond76.9%
Humanity's Last Exam9.9%
HLE-Diamond4.4%
MMMU-Pro56.8%
CritPt0.3%
CritPt v2 Hard0.0%
SciCode38.8%
FrontierCode 2.00.0%
Terminal-Bench Hard17.4%
Terminal-Bench 2.121.0%
Terminal-Bench 4.00.0%
Terminal-Bench-Science0.0%
τ²-Bench Telecom41.2%
τ³-Bench Banking5.0%
AutomationBench-AA1.0%
AA-AnalystAgent1.3%
AA-Briefcase241
GDPval-AA370
GDP.pdf0.0%
AA-LCR49.7%
MLCR-AA1.7%
IFBench48.2%
AA-Omniscience Index-31.0
AA-Omniscience Accuracy21.4%
AA-Omniscience Non-Hallucination33.5%

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 Small (latest)same series$0.15 / $0.60Mistral Small 3.2 24Bsame series—Mistral Small 3.1 24Bsame series—Gemma 4 31B ITcheaper alternative$0 / $0Nemotron 3 Ultra 550B A55Bcheaper alternative$0 / $0GPT-5 Nanocheaper alternative$0.05 / $0.40GLM-4.7-Flashcheaper alternative$0 / $0Mistral Small 3.2same lab$0.10 / $0.30
Data partly from models.dev (MIT) · Mistral official docs ↗