The State of Translation Automation 2026 · Languages

English to Spanish AI translation quality, cost and speed

What it takes to automate translation from English into European Spanish against 10 requirements in three levels: the solutions that meet them, what they cost and how fast they are, from the 2026 benchmark of 57 AI models and MT engines.

How good is AI translation from English to Spanish?

Key findings on English to Spanish translation

  1. 1Near missAnthropic Claude Opus 4.8, Cohere Command A Translate and DeepL Next Gen stay under the bar on weighted errors in European Spanish, but each has four major errors in the four texts, one more than general translation quality allows.See the data →
  2. 2Model typesOn the shortlist, one-step solutions on LLMs cost a median $3.74 per million characters, against $12.50 on MT engines and $18.75 on LLM-based MT engines, and the three solutions that meet the Basic requirements are all LLMs.See the data →
  3. 3Metrics vs LQAThe automatic metrics rank DeepL Next Gen first in European Spanish, but by Intento LQA its translations have 19.1 weighted errors per 1,000 words, against 9.9 for OpenAI GPT-5.6 Terra, third on COMET-Kiwi.See the data →

Find English to Spanish translation solutions for your requirements

Pick a quality level, or your own requirements, bars and constraints, to see the shortlisted solutions that meet them in European Spanish, with a column for each requirement you select. Under Optimize for, choose which of them comes first: the one with the fewest total errors, the lowest cost or the shortest time. Click a column heading to sort by it.

The shortlisted solutions on 11 models, with their Intento LQA results. Total errors per 1,000 words add the weighted general translation quality errors (minor 1, major 5, critical 10) to the missed glossary terms, sentences in the wrong register, tag errors and sentences that fail gender neutrality, each counted once. Formatting counts as met or missed and does not add to the total errors. The models that did not make the shortlist have automatic scores only: the shortlist. Gender neutrality: the Advanced requirements count the two best solutions of the language, with the gender-neutrality agent where they need it. The results of the other solutions are shown for information.

How we measure it: 11-01 · 11-02

Trade-offs in English to Spanish translation

Price vs quality

Two one-step solutions meet every Standard requirement in European Spanish at almost the same cost, $3.74 per million characters for OpenAI GPT-5.6 Terra and $3.87 for Google Gemini 3.1 Pro Preview. Translations from OpenAI GPT-5.6 Terra take 17 seconds per 1,000 words against 73 and have 20.1 total errors per 1,000 words against 24.6, so Optimize for selects OpenAI GPT-5.6 Terra for quality, cost and latency alike.

Constraints

Constraints narrow this quickly. Under the Standard requirements, a cost cap below $3.74 per million characters or a time limit below 17 seconds per 1,000 words removes both solutions, and the three real-time solutions on the shortlist, Microsoft Translator, Amazon Translate and DeepL Next Gen, all miss general translation quality and tag handling.

Agent effect

For the Advanced requirements, the gender-neutrality agent on OpenAI GPT-5.6 Terra raises the cost from $3.74 to $3.92 per million characters and the time from 17 to 24 seconds per 1,000 words, and its total errors fall from 20.1 to 6.3 per 1,000 words.

English to Spanish translation quality levels

Optimize for
Basic requirements★★★

Tier 1: simple solution + backup

Solutions that meet them
3 (3 one-step)
Simplest solution
one-step
Fewest errors

OpenAI GPT-5.6 Terra
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
20.1
Cost per million characters
$3.74
Seconds per 1,000 words
17
Lowest cost

xAI Grok 4.20 (reasoning)
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
26.6
Cost per million characters
$1.01
Seconds per 1,000 words
61
Shortest time

OpenAI GPT-5.6 Terra
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
20.1
Cost per million characters
$3.74
Seconds per 1,000 words
17
Standard requirements★★

Tier 1: simple solution + backup

Solutions that meet them
2 (2 one-step)
Simplest solution
one-step
Fewest errors

OpenAI GPT-5.6 Terra
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
20.1
Cost per million characters
$3.74
Seconds per 1,000 words
17
Lowest cost

OpenAI GPT-5.6 Terra
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
20.1
Cost per million characters
$3.74
Seconds per 1,000 words
17
Shortest time

OpenAI GPT-5.6 Terra
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
20.1
Cost per million characters
$3.74
Seconds per 1,000 words
17
Advanced requirements☆

Tier 3: one solution

Solutions that meet them
1 (0 one-step)
Simplest solution
with the gender-neutrality agent
Fewest errors

OpenAI GPT-5.6 Terra
one-step with the gender-neutrality agent

Total errors per 1,000 words
6.3
Cost per million characters
$3.92
Seconds per 1,000 words
24
Lowest cost

OpenAI GPT-5.6 Terra
one-step with the gender-neutrality agent

Total errors per 1,000 words
6.3
Cost per million characters
$3.92
Seconds per 1,000 words
24
Shortest time

OpenAI GPT-5.6 Terra
one-step with the gender-neutrality agent

Total errors per 1,000 words
6.3
Cost per million characters
$3.92
Seconds per 1,000 words
24

Optimize for selects one of the solutions that meet the level: the one with the fewest total errors, the lowest cost or the shortest time. Where no solution meets the level, the card shows the available solution with the fewest total errors and what it misses. Total errors count every requirement with countable errors.

Quality levelTierSolutionsSimplest solutionModelTotal errors per 1,000 wordsCost per million charactersSeconds per 1,000 words
BasicTier 1: simple solution + backup★★★one-stepxAI Grok 4.20 (reasoning)26.6$1.0161
StandardTier 1: simple solution + backup★★one-stepOpenAI GPT-5.6 Terra20.1$3.7417
AdvancedTier 3: one solution☆with the gender-neutrality agentOpenAI GPT-5.6 Terra6.3$3.9224

One solution per level: the simplest that meets the level, the cheapest of them where several are equally simple. ★ a one-step solution, ☆ a solution with more steps or the gender-neutrality agent.

Level step-up

Each level up removes one of the three solutions that meet the Basic requirements: tag handling the one on xAI Grok 4.20 (reasoning), with 99.2% of the tags kept, and gender neutrality the one on Google Gemini 3.1 Pro Preview, with 99.1% of the sentences about people gender-neutral. Under the Basic requirements, Optimize for selects xAI Grok 4.20 (reasoning) for cost, at $1.01 per million characters, and OpenAI GPT-5.6 Terra for quality and latency. Under the other two levels, it selects OpenAI GPT-5.6 Terra for all three.

How we measure it: 11-01 · 11-02 · 08-09

English to Spanish translation cost against errors

Chart data
Cost and weighted errors of the one-step solutions in European Spanish
ModelUS dollars per million charactersWeighted errors per 1,000 wordsGroup
xAI Grok 4.20 (reasoning)1.0115.2Misses a Standard requirement
AI21 Jamba Large 1.73.0423.7Misses a Standard requirement
OpenAI GPT-5.6 Terra3.749.9Meets every Standard requirement
Google Gemini 3.1 Pro Preview3.8714.8Meets every Standard requirement
Widn AI Tower Sugarloaf1035.3Misses a Standard requirement
Microsoft Translator1048.1Misses a Standard requirement
Anthropic Claude Opus 4.814.0614.2Misses a Standard requirement
Amazon Translate1561.3Misses a Standard requirement
DeepL Next Gen27.519.1Misses a Standard requirement
Price vs quality

Among the one-step solutions, a higher price does not come with fewer errors. Anthropic Claude Opus 4.8 costs $14.06 per million characters and has 14.2 weighted errors per 1,000 words, Amazon Translate $15 and 61.3, and DeepL Next Gen $27.50 and 19.1. The four cheapest, all LLMs from $1.01 to $3.87, span 9.9 to 23.7 weighted errors per 1,000 words.

How we measure it: 11-01 · 10-10

English to Spanish translation requirements

Chart data
Errors per 1,000 words of the one-step solutions in European Spanish, median, by requirement
RequirementMedian errors per 1,000 words
General translation quality19.1
Gender neutrality9.8
Tag handling2.3
Tone of voice2.1
Terminology0

The requirements with countable errors. Every solution with its result on each requirement: the explorer.

Near miss

Terminology and tags are missed by small margins: Anthropic Claude Opus 4.8 and Cohere Command A Translate get 98.7% and 98.8% of the glossary terms right, and xAI Grok 4.20 (reasoning), AI21 Jamba Large 1.7 and DeepL Next Gen keep 99.2%, 98.6% and 96.5% of the tags. Tone of voice separates the MT engines: Microsoft Translator has 4.3% of its sentences in the tone of voice, Amazon Translate 74.5% and Elia Machine Translation System 76.6%, while every LLM solution reaches 100%.

How we measure it: 09-09 · 09-10 · 09-11 · 09-12 · 09-18

AI models and MT engines for English to Spanish translation

Chart data
Weighted errors per 1,000 words of the shortlisted one-step solutions in European Spanish
ModelWeighted errors per 1,000 words of the shortlisted one-step solutions in European SpanishClass
Elia Machine Translation System72.5MT engines
Amazon Translate61.3MT engines
Microsoft Translator48.1MT engines
Widn AI Tower Sugarloaf35.3LLM-based MT engines
AI21 Jamba Large 1.723.7LLMs
Cohere Command A Translate19.1LLMs
DeepL Next Gen19.1LLM-based MT engines
xAI Grok 4.20 (reasoning)15.2LLMs
Google Gemini 3.1 Pro Preview14.8LLMs
Anthropic Claude Opus 4.814.2LLMs
OpenAI GPT-5.6 Terra9.9LLMs

No translation into European Spanish: Widn AI Tower Vesuvius (cut off part-way).

Model types

The Intento LQA results group by model type in European Spanish: the six LLM solutions on the shortlist have 9.9 to 23.7 weighted errors per 1,000 words, the two LLM-based MT engines 19.1 and 35.3, and the three MT engines 48.1 to 72.5. All six shortlisted LLMs support the language by our measured criteria, and three of them meet general translation quality.

How we measure it: 00-04 · 08-15 · 23-01

How we shortlisted English to Spanish translation solutions

Chart data
Ranks of the models on the two automatic metrics, European Spanish
ModelCOMET-KiwiRank on COMET-KiwiMetricXRank on MetricXShortlistedShortlist place
Google Gemini 3.5 Flash0.80791.3314NoNot shortlisted
Anthropic Claude Fable 50.807101.3520NoNot shortlisted
Anthropic Claude Sonnet 4.60.807111.4129NoNot shortlisted
MoonshotAI Kimi K2.50.807121.3213NoNot shortlisted
Anthropic Claude Sonnet 50.807131.4127NoNot shortlisted
OpenAI GPT-5.50.807141.3112NoNot shortlisted
Mistral Large 30.806151.3416NoNot shortlisted
DeepSeek V3.20.806161.3622NoNot shortlisted
OpenAI GPT-4.10.806171.3010NoNot shortlisted
z.ai GLM-50.805181.3621NoNot shortlisted
Meta Llama 4 Maverick0.804191.4128NoNot shortlisted
Google Gemma 40.803221.3519NoNot shortlisted
ModernMT Enterprise Edition0.803231.309NoNot shortlisted
Anthropic Claude Haiku 4.50.802241.3725NoNot shortlisted
MiniMax M2.50.801251.4633NoNot shortlisted
Swiss AI Apertus-8B-25090.801271.3623NoNot shortlisted
Translated Lara 20.800281.6738NoNot shortlisted
Supertext LLM0.800291.4331NoNot shortlisted
OpenAI gpt-oss-120b0.799301.3624NoNot shortlisted
Mistral Medium 3.10.796311.4026NoNot shortlisted
Google Translation LLM0.796321.7440NoNot shortlisted
SpeakLeash Bielik-11B-v3.0-Instruct0.795331.4332NoNot shortlisted
Pangeanic PangeaMT0.793341.4230NoNot shortlisted
Tilde TildeOpen LLM engine0.792351.4935NoNot shortlisted
Alibaba Qwen3-235B-A22B0.791361.3518NoNot shortlisted
SYSTRAN PNMT API0.790371.4734NoNot shortlisted
NiuTrans0.789381.6037NoNot shortlisted
Yandex Cloud Translate v20.786391.5636NoNot shortlisted
Oracle Cloud Language Service0.774401.6839NoNot shortlisted
Google Cloud Advanced Translation API0.771412.3645NoNot shortlisted
iFlyTek MT0.771421.8741NoNot shortlisted
Tencent Cloud TMT0.765431.9842NoNot shortlisted
Naver Papago NMT Commercial0.764442.0343NoNot shortlisted
Alibaba Machine Translation API0.759452.0544NoNot shortlisted
Cohere Tiny Aya0.751462.3846NoNot shortlisted
AppTek NMT0.699473.3247NoNot shortlisted
Widn AI Tower Anthill0.4854810.1448NoNot shortlisted
DeepL Next Gen0.81811.141YesTop 8 by both metrics
xAI Grok 4.20 (reasoning)0.81121.3011YesTop 8 by COMET-Kiwi
OpenAI GPT-5.6 Terra0.81131.298YesTop 8 by both metrics
Google Gemini 3.1 Pro Preview0.81041.274YesTop 8 by both metrics
Cohere Command A Translate0.81051.3315YesTop 8 by COMET-Kiwi
Anthropic Claude Opus 4.80.80961.3417YesTop 8 by COMET-Kiwi
Amazon Translate0.80971.287YesTop 8 by both metrics
Widn AI Tower Sugarloaf0.80881.286YesTop 8 by both metrics
Microsoft Translator0.804201.222YesTop 8 by MetricX
AI21 Jamba Large 1.70.803211.275YesTop 8 by MetricX
Elia Machine Translation System0.801261.243YesTop 8 by MetricX
ModelSolution complexityCOMET-Kiwi rankCOMET-KiwiMetricX rankMetricXShortlist
DeepL Next GenAdjusted NMT10.81811.14Top 8 by both metrics
xAI Grok 4.20 (reasoning)Adjusted LLM20.811111.30Top 8 by COMET-Kiwi
OpenAI GPT-5.6 TerraAdjusted LLM30.81181.29Top 8 by both metrics
Google Gemini 3.1 Pro PreviewAdjusted LLM40.81041.27Top 8 by both metrics
Cohere Command A TranslateAdjusted LLM50.810151.33Top 8 by COMET-Kiwi
Anthropic Claude Opus 4.8Adjusted LLM60.809171.34Top 8 by COMET-Kiwi
Amazon TranslateOff-the-shelf NMT70.80971.28Top 8 by both metrics
Widn AI Tower SugarloafAdjusted NMT80.80861.28Top 8 by both metrics
Google Gemini 3.5 FlashAdjusted LLM90.807141.33Not shortlisted
Anthropic Claude Fable 5Adjusted LLM100.807201.35Not shortlisted
Anthropic Claude Sonnet 4.6Adjusted LLM110.807291.41Not shortlisted
MoonshotAI Kimi K2.5Adjusted LLM120.807131.32Not shortlisted
Anthropic Claude Sonnet 5Adjusted LLM130.807271.41Not shortlisted
OpenAI GPT-5.5Adjusted LLM140.807121.31Not shortlisted
Mistral Large 3Adjusted LLM150.806161.34Not shortlisted
DeepSeek V3.2Adjusted LLM160.806221.36Not shortlisted
OpenAI GPT-4.1Adjusted LLM170.806101.30Not shortlisted
z.ai GLM-5Adjusted LLM180.805211.36Not shortlisted
Meta Llama 4 MaverickAdjusted LLM190.804281.41Not shortlisted
Microsoft TranslatorOff-the-shelf NMT200.80421.22Top 8 by MetricX
AI21 Jamba Large 1.7Adjusted LLM210.80351.27Top 8 by MetricX
Google Gemma 4Adjusted LLM220.803191.35Not shortlisted
ModernMT Enterprise EditionAdjusted NMT230.80391.30Not shortlisted
Anthropic Claude Haiku 4.5Adjusted LLM240.802251.37Not shortlisted
MiniMax M2.5Adjusted LLM250.801331.46Not shortlisted
Elia Machine Translation SystemOff-the-shelf NMT260.80131.24Top 8 by MetricX
Swiss AI Apertus-8B-2509Adjusted LLM270.801231.36Not shortlisted
Translated Lara 2Adjusted NMT280.800381.67Not shortlisted
Supertext LLMOff-the-shelf NMT290.800311.43Not shortlisted
OpenAI gpt-oss-120bAdjusted LLM300.799241.36Not shortlisted
Mistral Medium 3.1Adjusted LLM310.796261.40Not shortlisted
Google Translation LLMOff-the-shelf NMT320.796401.74Not shortlisted
SpeakLeash Bielik-11B-v3.0-InstructAdjusted LLM330.795321.43Not shortlisted
Pangeanic PangeaMTOff-the-shelf NMT340.793301.42Not shortlisted
Tilde TildeOpen LLM engineAdjusted NMT350.792351.49Not shortlisted
Alibaba Qwen3-235B-A22BAdjusted LLM360.791181.35Not shortlisted
SYSTRAN PNMT APIAdjusted NMT370.790341.47Not shortlisted
NiuTransOff-the-shelf NMT380.789371.60Not shortlisted
Yandex Cloud Translate v2Adjusted NMT390.786361.56Not shortlisted
Oracle Cloud Language ServiceOff-the-shelf NMT400.774391.68Not shortlisted
Google Cloud Advanced Translation APIAdjusted NMT410.771452.36Not shortlisted
iFlyTek MTOff-the-shelf NMT420.771411.87Not shortlisted
Tencent Cloud TMTOff-the-shelf NMT430.765421.98Not shortlisted
Naver Papago NMT CommercialOff-the-shelf NMT440.764432.03Not shortlisted
Alibaba Machine Translation APIOff-the-shelf NMT450.759442.05Not shortlisted
Cohere Tiny AyaOff-the-shelf LLM460.751462.38Not shortlisted
AppTek NMTOff-the-shelf NMT470.699473.32Not shortlisted
Widn AI Tower AnthillAdjusted NMT480.4854810.14Not shortlisted
OpenAI GPT-5.6 LunaAdjusted LLM—0.807—1.36Same model family as OpenAI GPT-5.6 Terra
OpenAI GPT-5.6 SolAdjusted LLM—0.806—1.33Same model family as OpenAI GPT-5.6 Terra
xAI Grok 4.20 (non-reasoning)Adjusted LLM—0.803—1.40Same model family as xAI Grok 4.20 (reasoning)

Every model we got a translation from, as the shortlist compared it: off the shelf, without context in the request, or adjusted, with it. A model that did not make the shortlist is shown adjusted, unless off the shelf it scores higher by 0.03 or more on COMET-Kiwi. Ranks among 48 models, with each model family counted once, and the top 8 of each metric in bold. Higher is better on COMET-Kiwi, lower on MetricX. The scores rank models within this language and are not a requirement result.

Metrics vs LQA

The metrics and Intento LQA order the shortlisted solutions differently. DeepL Next Gen, first on both metrics, misses general translation quality, tag handling and formatting, while xAI Grok 4.20 (reasoning), on the shortlist through COMET-Kiwi alone and 11th on MetricX, meets general translation quality. Microsoft Translator, second on MetricX and 20th on COMET-Kiwi, has 48.1 weighted errors per 1,000 words. The models nearest the cut are Google Gemini 3.5 Flash, 9th on COMET-Kiwi, and ModernMT Enterprise Edition, 9th on MetricX.

How we measure it: 00-08 · 00-09

English to Spanish translation cost

Quality levelCheapest solution that meets every requirement of the levelCost per million characters
BasicxAI Grok 4.20 (reasoning), one-step LLM adjusted for Standard requirements$1.01
StandardOpenAI GPT-5.6 Terra, one-step LLM adjusted for Standard requirements$3.74
AdvancedOpenAI GPT-5.6 Terra, one-step with the gender-neutrality agent$3.92
Chart data
Cost of the shortlisted one-step solutions in European Spanish
ModelUS dollars per million charactersClass
xAI Grok 4.20 (reasoning)$1.01LLMs
AI21 Jamba Large 1.7$3.04LLMs
OpenAI GPT-5.6 Terra$3.74LLMs
Google Gemini 3.1 Pro Preview$3.87LLMs
Widn AI Tower Sugarloaf$10LLM-based MT engines
Microsoft Translator$10MT engines
Anthropic Claude Opus 4.8$14.06LLMs
Amazon Translate$15MT engines
DeepL Next Gen$27.5LLM-based MT engines
Chart data
Tokens per character of the translation into European Spanish, by model
ModelTokens per characterClass
Anthropic Claude Opus 4.80.417LLMs
Anthropic Claude Sonnet 50.417LLMs
Anthropic Claude Fable 50.416LLMs
Mistral Medium 3.10.337LLMs
Mistral Large 30.326LLMs
Anthropic Claude Haiku 4.50.309LLMs
Anthropic Claude Sonnet 4.60.308LLMs
MoonshotAI Kimi K2.50.292LLMs
AI21 Jamba Large 1.70.274LLMs
Alibaba Qwen3-235B-A22B0.273Open-weight LLMs
Widn AI Tower Anthill0.272LLM-based MT engines
Widn AI Tower Sugarloaf0.271LLM-based MT engines
DeepSeek V3.20.268LLMs
z.ai GLM-50.255LLMs
Cohere Command A Translate0.248LLMs
Google Gemini 3.1 Pro Preview0.242LLMs
Google Gemini 3.5 Flash0.242LLMs
Cohere Tiny Aya0.241LLMs
Google Gemma 40.241Open-weight LLMs
OpenAI GPT-4.10.238LLMs
OpenAI GPT-5.50.237LLMs
xAI Grok 4.20 (non-reasoning)0.236LLMs
xAI Grok 4.20 (reasoning)0.236LLMs
OpenAI GPT-5.6 Terra0.236LLMs
OpenAI GPT-5.6 Luna0.235LLMs
OpenAI GPT-5.6 Sol0.235LLMs
MiniMax M2.50.214LLMs
Cost driver

With billing on tokens, the tokenizers produce a median 0.26 tokens per character of European Spanish, 1.1 times their count for English, and the tokenizer of Anthropic Claude Opus 4.8 produces 1.9 times as many tokens per character as the tokenizer of MiniMax M2.5.

How we measure it: 10-01 · 10-03 · 05-02

English to Spanish translation speed

Quality levelFastest solution that meets every requirement of the levelSeconds per 1,000 words
BasicOpenAI GPT-5.6 Terra, one-step LLM adjusted for Standard requirements17
StandardOpenAI GPT-5.6 Terra, one-step LLM adjusted for Standard requirements17
AdvancedOpenAI GPT-5.6 Terra, one-step with the gender-neutrality agent24
Chart data
Time to translate 1,000 words into European Spanish, by model
ModelSeconds per 1,000 wordsClass
Microsoft Translator0.2MT engines
Pangeanic PangeaMT0.5MT engines
Google Cloud Advanced Translation API1MT engines
Translated Lara 21.1LLM-based MT engines
ModernMT Enterprise Edition1.2MT engines
Google Translation LLM1.2LLM-based MT engines
OpenAI gpt-oss-120b1.3Open-weight LLMs
SYSTRAN PNMT API1.3MT engines
Amazon Translate1.5MT engines
DeepL Next Gen2LLM-based MT engines
Yandex Cloud Translate v22.9MT engines
Elia Machine Translation System3MT engines
Oracle Cloud Language Service4.1MT engines
Widn AI Tower Anthill7.8LLM-based MT engines
Tilde TildeOpen LLM engine10LLM-based MT engines
Google Gemini 3.5 Flash13LLMs
OpenAI GPT-5.6 Luna13LLMs
Supertext LLM14LLM-based MT engines
OpenAI GPT-5.6 Terra17LLMs
OpenAI GPT-4.118LLMs
Cohere Tiny Aya18LLMs
xAI Grok 4.20 (non-reasoning)19LLMs
Alibaba Machine Translation API22MT engines
Anthropic Claude Haiku 4.523LLMs
Cohere Command A Translate27LLMs
Mistral Medium 3.131LLMs
MiniMax M2.533LLMs
Anthropic Claude Sonnet 534LLMs
Mistral Large 337LLMs
DeepSeek V3.237LLMs
AI21 Jamba Large 1.740LLMs
OpenAI GPT-5.541LLMs
Meta Llama 4 Maverick41Open-weight LLMs
OpenAI GPT-5.6 Sol41LLMs
Anthropic Claude Opus 4.843LLMs
Widn AI Tower Sugarloaf43LLM-based MT engines
Swiss AI Apertus-8B-250945Open-weight LLMs
Anthropic Claude Sonnet 4.651LLMs
iFlyTek MT52MT engines
SpeakLeash Bielik-11B-v3.0-Instruct54Open-weight LLMs
NiuTrans54MT engines
Tencent Cloud TMT56MT engines
xAI Grok 4.20 (reasoning)61LLMs
Google Gemini 3.1 Pro Preview73LLMs
Anthropic Claude Fable 573LLMs
Naver Papago NMT Commercial73MT engines
Alibaba Qwen3-235B-A22B92Open-weight LLMs
AppTek NMT95MT engines
Google Gemma 4170Open-weight LLMs
z.ai GLM-5189LLMs
MoonshotAI Kimi K2.5237LLMs

Real-time models are those the shortlist counts as real-time, under 2 seconds per text across the languages we timed. The chart shows the time of each model into European Spanish.

Real-time

Nine of the ten real-time models in European Spanish are MT engines or LLM-based MT engines, and the other one is the open-weight OpenAI gpt-oss-120b, which did not make the shortlist. The shortlisted one-step LLM solutions take from 17 to 73 seconds per 1,000 words, OpenAI GPT-5.6 Terra the least and Google Gemini 3.1 Pro Preview the most.

How we measure it: 12-01 · 12-02 · 12-09

Tone of voice and gender-neutral English to Spanish translation

Tone of voice
Formal register, with usted
Gender-neutrality strategy
neutral nouns + rephrasing; doubling only when unavoidable; no glyphs
Options in the language
generic masculine, paired (doubling), typographic (*,@,x,-e), neutral nouns+rephrasing
Default gender
None where the source leaves the gender open
The two best solutionsSolutionSentences about people that are gender-neutralMeets every Advanced requirementCost per million charactersSeconds per 1,000 words
OpenAI GPT-5.6 Terraone-step, with the gender-neutrality agent100%✓$3.9224
Google Gemini 3.1 Pro Previewthree-step, with the gender-neutrality agent99.1%✗$989252
Agent effect

For gender neutrality in European Spanish we check neutral nouns and rephrasing, with paired forms only where they cannot be avoided. The gender-neutrality agent works with both of the two best solutions. With it, the three-step solution (translation + LQA + repair) on Google Gemini 3.1 Pro Preview costs $989 per million characters and takes 252 seconds per 1,000 words, against $3.92 and 24 seconds for the one-step solution on OpenAI GPT-5.6 Terra.

How we measure it: 02-05 · 02-06 · 08-11 · 09-18

How confident are we in measuring English to Spanish translation quality?

Confidence

Both pairs of automatic metrics agree in European Spanish, and the three LQA judges agree on a median of 94.9% of the sentences of the shortlisted solutions, so the results on this page rest on measurement we can rely on.

How we measure it: 00-07 · 07-05 · 09-02

Automate translation to meet your requirements

See Intento Language Hub in action.

Book a demo
Leave feedback