Multilingual AI models: all 239 handle more than English

Prices from official vendor price lists · October 2026
All 239 models in the rating work beyond English, including Arabic. The difference is how efficiently they tokenise it. Output prices run from $0.059 to $180 per 1M tokens across the 199 models with a paid per-token price; 80 have a free tier and 88 ship open weights.
Multilingual
Services found: 239
#AI modelCapabilitiesAPI $/1M (in / out)Context windowPrice / moOutput $/1MOverviewVisit site
101

Command R7B

Cohere
Text & chatMultilingualProprietary API
128K tokensAPI onlyView
102

North Mini Code

Cohere
Text & chatCodingLow latency
256K tokensAPI onlyView
103

Jamba2 Mini

AI21 Labs
Text & chatReasoningTool calling
256K tokensAPI onlyView
104

Jamba2 3B

AI21 Labs
Text & chatReasoningTool calling
256K tokensAPI onlyView
105

Nova 2 Lite

Amazon
Text & chatVisionVideo
$0.3 / $2.5256K tokensAPI only2.5View
106

Nova Premier

Amazon
Text & chatVisionVideo
API onlyView
107

Nova Pro

Amazon
Text & chatVisionVideo
300K tokensAPI onlyView
108

Nova Lite

Amazon
Text & chatVisionVideo
300K tokensAPI onlyView
109

Nova Micro

Amazon
Text & chatLow latencyMultilingual
128K tokensAPI onlyView
110

Nova 2 Sonic

Amazon
Text & chatAudioRealtime voice
API onlyView
111

Nova Sonic

Amazon
Text & chatAudioRealtime voice
API onlyView
112

Phi-4 Mini Instruct

Microsoft
Text & chatReasoningTool calling
$0.075 / $0.30131.1K tokensAPI only0.3View
113

Phi-4 Reasoning

Microsoft
Text & chatReasoningTool calling
32.8K tokensAPI onlyView
114

Phi-4 Mini Reasoning

Microsoft
Text & chatReasoningTool calling
128K tokensAPI onlyView
115

Qwen3.8 Max

Alibaba
Text & chatReasoningAdjustable thinking
$2 / $61M tokensAPI only6View
116

Qwen3 VL Plus

Alibaba
Text & chatVisionReasoning
$0.2 / $1.6256K tokensAPI only1.6View
117

Qwen3 VL Flash

Alibaba
Text & chatVisionReasoning
$0.05 / $0.4256K tokensAPI only0.4View
118

ERNIE 5.1

Baidu
Text & chatReasoningTool calling
$0.56 / $2.53API only2.53View
119

ERNIE 5.0

Baidu
Text & chatVisionReasoning
$1.4 / $5.6119K tokensAPI only5.6View
120

ERNIE 4.5 Turbo VL

Baidu
Text & chatVisionReasoning
128K tokensAPI onlyView

Browse by need

Reasoning models169 modelsCoding assistants17 modelsVision and images156 modelsLong context 1M+82 modelsFree tier available80 modelsOpen weights88 models

Multilingual capability is now table stakes: all 239 models in this rating handle languages beyond English at production quality, Arabic included. The interesting question is no longer whether a model speaks your language but how efficiently it tokenises it.

Non-Latin scripts consume more tokens per word than English, so the same Arabic text costs more to process than its English equivalent. That makes the cheap end of this list - output from $0.25 per million - disproportionately attractive for high-volume multilingual work, and it makes prompt caching worth checking if your system prompt is long.

Tokenisation, not fluencyEvery model speaks the language; the difference is how many tokens it burns doing it.
Arabic costs moreNon-Latin scripts inflate token counts, so the same text bills higher than its English version.
Caching helps most hereA long multilingual system prompt is exactly what prompt caching was designed for.

Models with this capability

The 239 models below all support this capability. Prices are input and output per 1M tokens from the vendor official price list; for dirhams multiply by 3.6725. Tap a card to open the model page.

Claude Fable 5.1
#1
Anthropic
in $10out $501M tokens
GPT-6 Astra
#2
OpenAI
in $10out $501.05M tokens
GPT-6.1 Sol
#3
OpenAI
in $2out $101.05M tokens
GPT-6 Sol
#4
OpenAI
in $2out $101.05M tokens
GPT-6 Luna
#5
OpenAI
in $0.1out $0.51.05M tokens
Claude Opus 5.5
#6
Anthropic
in $4out $201M tokens
Claude Sonnet 5.5
#7
Anthropic
in $2out $101M tokens
Grok 4.7
#8
xAI
in $2out $6500K tokens

How to choose

  • Measure cost on real text in your language, not on English benchmarks.
  • Check output quality on dialect and register, not just on literal accuracy.
  • Prefer models with a free tier so you can test on your own corpus before paying.
  • For regulated translation, compare against a dedicated translation model.

Common mistakes

Assuming English pricing appliesA budget that works in English can overrun by half again in Arabic.
Skipping native reviewFluent output hides subtle register errors that only a native reader catches.

Who this suits

Gulf-facing productsArabic and English in one interface, with pricing that survives both.
Global supportOne model covering the whole ticket queue instead of per-language routing.
Content teamsDrafting in one language and adapting rather than translating.

Related filters

TranslationFaithful translation with terminology control
Prompt caching147 models with cheap repeated context
Frontier tier41 flagship models, the best each vendor sells
Free tier80 models with a free API allowance
All 239 modelsThe full rating with filters and prices

Frequently asked questions

Does Arabic cost more than English?

Yes. Non-Latin scripts use more tokens per word, so the same text bills higher. Budget models and prompt caching offset it.

Which models are strongest in Arabic?

The frontier tier handles it best, and regional families — Qwen, GLM, Falcon — are trained with heavier non-English weighting.

Is a translation model better than a multilingual one?

For contracts and regulated text, yes. For everyday work a general mid-tier model is cheaper and good enough.

Which languages are weakest?

Low-resource languages remain uneven across all vendors. Test rather than assume.

Does a bigger model speak better?

Usually yes on nuance, but the gap narrows for straightforward everyday text.

Frequently searched

By capabilityReasoningCodingVisionLong context 1M+Free tierOpen weightsLow latencyRealtime voiceImage generation
By vendorAnthropic (13)OpenAI (30)Google DeepMind (17)xAI (8)Meta (6)DeepSeek (9)Mistral AI (11)Alibaba (Qwen) (34)Z.ai (12)Moonshot AI (5)
Popular modelsClaude Fable 5.1GPT-6 AstraClaude Opus 5Claude Fable 5GPT-5.6 SolGLM 5.3Grok 4.6Kimi K3GPT-5.6 Terra
By deliveryBatch APIPrompt cachingPriority tierFrontier tierProprietary APITool callingStructured JSONWeb searchMultilingual

Ratings 0

Avg. score 0

Rate this page

Help us improve