The State of Translation Automation 2026 · Languages

English to Vietnamese AI translation quality, cost and speed

What it takes to automate translation from English into Vietnamese 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 Vietnamese?

Key findings on English to Vietnamese translation

  1. 1Near missThe solutions on Google Gemini 3.5 Flash and Google Gemini 3.1 Pro Preview meet tone of voice, tag handling and formatting in Vietnamese, and cost $3.25 and $4.37 per million characters. Their translations have 22.1 weighted errors per 1,000 words, more than general translation quality allows.See the data →
  2. 2Real-timeThree real-time (<2 sec) solutions made the shortlist, on DeepL Next Gen, Translated Lara 2 and Google Translation LLM. None of them meets general translation quality, the one Basic requirement: their translations have 23.4, 31.3 and 69.9 weighted errors per 1,000 words.See the data →
  3. 3Metrics vs LQAThe solution on DeepL Next Gen, first on both automatic metrics in Vietnamese, misses general translation quality, tone of voice, tag handling and formatting. The two solutions that meet every requirement, on Anthropic Claude Opus 4.8 and Anthropic Claude Fable 5, made the shortlist through COMET-Kiwi alone.See the data →

Find English to Vietnamese 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 Vietnamese, 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 12 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 Vietnamese translation

Price vs quality

The two solutions that meet every Standard requirement in Vietnamese are one-step solutions on Anthropic LLMs. Translations from Anthropic Claude Fable 5 have 13.8 total errors per 1,000 words, as many as from Anthropic Claude Opus 4.8, and cost $33.56 per million characters against $16.82. They take 68 and 67 seconds per 1,000 words. Optimizing for quality marks both, as they tie on total errors, and optimizing for cost or latency selects the solution on Anthropic Claude Opus 4.8.

Constraints

A cost cap of $16.81 per million characters or less removes both solutions that meet every Standard requirement, and a cap between $16.82 and $33.56 keeps only the one on Anthropic Claude Opus 4.8. A time limit of 66 seconds per 1,000 words or less removes both. A real-time limit removes every solution that meets general translation quality. A limit of one step changes nothing, because both solutions are one-step solutions.

Agent effect

The Advanced requirements add gender neutrality. In Vietnamese, translations from both solutions that meet every Standard requirement have 100% of the sentences about people gender-neutral without the gender-neutrality agent. So meeting the Advanced requirements costs the same $16.82 per million characters and takes the same 67 seconds per 1,000 words as meeting the Standard ones.

English to Vietnamese translation quality levels

Optimize for
Basic requirements★★

Tier 1: simple solution + backup

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

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Tied with Anthropic Claude Fable 5 (one-step LLM adjusted for Standard requirements).

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Lowest cost

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Shortest time

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Standard requirements★★

Tier 1: simple solution + backup

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

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Tied with Anthropic Claude Fable 5 (one-step LLM adjusted for Standard requirements).

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Lowest cost

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Shortest time

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Advanced requirements★★

Tier 1: simple solution + backup

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

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Tied with Anthropic Claude Fable 5 (one-step LLM adjusted for Standard requirements).

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Lowest cost

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67
Shortest time

Anthropic Claude Opus 4.8
one-step LLM adjusted for Standard requirements

Total errors per 1,000 words
13.8
Cost per million characters
$16.82
Seconds per 1,000 words
67

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-stepAnthropic Claude Opus 4.813.8$16.8267
StandardTier 1: simple solution + backup★★one-stepAnthropic Claude Opus 4.813.8$16.8267
AdvancedTier 1: simple solution + backup★★one-stepAnthropic Claude Opus 4.813.8$16.8267

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

In Vietnamese, each level up keeps both solutions that meet the level below. The one-step solutions on Anthropic Claude Fable 5 and Anthropic Claude Opus 4.8 meet general translation quality, the one Basic requirement, then tone of voice, tag handling and formatting, then gender neutrality. Each level takes one step, without the gender-neutrality agent. Optimize for makes the same selection at every level: for quality both solutions, which tie on total errors, and for cost and latency the solution on Anthropic Claude Opus 4.8.

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

English to Vietnamese translation cost against errors

Chart data
Cost and weighted errors of the one-step solutions in Vietnamese
ModelUS dollars per million charactersWeighted errors per 1,000 wordsGroup
Google Gemma 40.1335.3Misses a Standard requirement
Mistral Large 31.0946.1Misses a Standard requirement
Google Gemini 3.5 Flash3.2522.1Misses a Standard requirement
Google Gemini 3.1 Pro Preview4.3722.1Misses a Standard requirement
Anthropic Claude Sonnet 56.6826Misses a Standard requirement
Anthropic Claude Sonnet 4.68.6725.7Misses a Standard requirement
Anthropic Claude Opus 4.816.8213.8Meets every Standard requirement
Google Translation LLM21.3769.9Misses a Standard requirement
Translated Lara 224.9931.3Misses a Standard requirement
DeepL Next Gen27.523.4Misses a Standard requirement
Anthropic Claude Fable 533.5613.8Meets every Standard requirement
Price vs quality

In Vietnamese, no solution that costs more than the one on Anthropic Claude Opus 4.8, at $16.82 per million characters, has fewer weighted errors than its 13.8 per 1,000 words. Translations from Anthropic Claude Fable 5, at $33.56 per million characters, have as many weighted errors, with one major error against none. The solutions on the three LLM-based MT engines cost $21.37 to $27.50 per million characters and have 23.4 to 69.9 weighted errors per 1,000 words.

How we measure it: 11-01 · 10-10

English to Vietnamese translation requirements

Chart data
Errors per 1,000 words of the one-step solutions in Vietnamese, median, by requirement
RequirementMedian errors per 1,000 words
General translation quality25.9
Tone of voice0.9
Tag handling0.7
Gender neutrality0

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

Bottleneck

General translation quality is the requirement the most one-step solutions miss in Vietnamese, ten of the 12, with 22.1 to 69.9 weighted errors per 1,000 words. The one-step solutions on Anthropic Claude Opus 4.8 and Anthropic Claude Fable 5 meet every Standard requirement, so two-step solutions (translation + post-edit) and three-step solutions (translation + LQA + repair) do not change the tier, tier 1 (simple solution + backup).

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

AI models and MT engines for English to Vietnamese translation

Chart data
Weighted errors per 1,000 words of the shortlisted one-step solutions in Vietnamese
ModelWeighted errors per 1,000 words of the shortlisted one-step solutions in VietnameseClass
Google Translation LLM69.9LLM-based MT engines
Mistral Large 346.1LLMs
Google Gemma 435.3Open-weight LLMs
Cohere Command A Translate32LLMs
Translated Lara 231.3LLM-based MT engines
Anthropic Claude Sonnet 526LLMs
Anthropic Claude Sonnet 4.625.7LLMs
DeepL Next Gen23.4LLM-based MT engines
Google Gemini 3.1 Pro Preview22.1LLMs
Google Gemini 3.5 Flash22.1LLMs
Anthropic Claude Opus 4.813.8LLMs
Anthropic Claude Fable 513.8LLMs

No translation into Vietnamese: Widn AI Tower Anthill (far too short, wrong language), Widn AI Tower Vesuvius (far too short, wrong language).

Model types

In Vietnamese, translations from the shortlisted LLMs have 13.8 to 46.1 weighted errors per 1,000 words, from the open-weight Google Gemma 4 35.3, and from the three LLM-based MT engines 23.4 to 69.9. Translations from DeepL Next Gen, the LLM-based MT engine with the fewest weighted errors, have 1.7 times as many weighted errors as translations from Anthropic Claude Opus 4.8. All nine shortlisted LLMs, Google Gemma 4 included, support Vietnamese by our measured criteria, and the solutions on two of them meet every requirement.

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

How we shortlisted English to Vietnamese translation solutions

Chart data
Ranks of the models on the two automatic metrics, Vietnamese
ModelCOMET-KiwiRank on COMET-KiwiMetricXRank on MetricXShortlistedShortlist place
OpenAI GPT-5.6 Luna0.79791.179NoNot shortlisted
OpenAI GPT-5.50.794101.2616NoNot shortlisted
OpenAI GPT-4.10.794131.1911NoNot shortlisted
MoonshotAI Kimi K2.50.793141.2213NoNot shortlisted
z.ai GLM-50.792151.1912NoNot shortlisted
DeepSeek V3.20.791161.2717NoNot shortlisted
xAI Grok 4.20 (non-reasoning)0.789171.3119NoNot shortlisted
Mistral Medium 3.10.788181.2515NoNot shortlisted
Swiss AI Apertus-8B-25090.783211.5328NoNot shortlisted
Meta Llama 4 Maverick0.782221.2818NoNot shortlisted
Microsoft Translator0.780231.3522NoNot shortlisted
Anthropic Claude Haiku 4.50.779241.3825NoNot shortlisted
MiniMax M2.50.778251.3623NoNot shortlisted
Alibaba Qwen3-235B-A22B0.777261.3824NoNot shortlisted
ModernMT Enterprise Edition0.776271.3321NoNot shortlisted
OpenAI gpt-oss-120b0.776281.4126NoNot shortlisted
Google Cloud Advanced Translation API0.770292.0834NoNot shortlisted
Amazon Translate0.769301.4827NoNot shortlisted
SYSTRAN PNMT API0.753311.6630NoNot shortlisted
Tencent Cloud TMT0.748321.8031NoNot shortlisted
Yandex Cloud Translate v20.738331.9332NoNot shortlisted
Naver Papago NMT Commercial0.733341.9933NoNot shortlisted
Alibaba Machine Translation API0.717352.2535NoNot shortlisted
AI21 Jamba Large 1.70.707363.1936NoNot shortlisted
Cohere Tiny Aya0.694373.5137NoNot shortlisted
AppTek NMT0.650383.7938NoNot shortlisted
DeepL Next Gen0.80211.071YesTop 8 by both metrics
Google Gemini 3.5 Flash0.80121.134YesTop 8 by both metrics
Anthropic Claude Sonnet 50.80131.178YesTop 8 by both metrics
Anthropic Claude Fable 50.80041.1710YesTop 8 by COMET-Kiwi
Google Gemini 3.1 Pro Preview0.79951.145YesTop 8 by both metrics
Anthropic Claude Opus 4.80.79861.2214YesTop 8 by COMET-Kiwi
Cohere Command A Translate0.79871.123YesTop 8 by both metrics
Anthropic Claude Sonnet 4.60.79781.146YesTop 8 by both metrics
Google Gemma 40.794111.167YesTop 8 by MetricX
Mistral Large 30.794121.112YesTop 8 by MetricX
Translated Lara 20.787191.3220YesReal-time place
Google Translation LLM0.787201.6429YesReal-time place
ModelSolution complexityCOMET-Kiwi rankCOMET-KiwiMetricX rankMetricXShortlist
DeepL Next GenOff-the-shelf NMT10.80211.07Top 8 by both metrics
Google Gemini 3.5 FlashAdjusted LLM20.80141.13Top 8 by both metrics
Anthropic Claude Sonnet 5Adjusted LLM30.80181.17Top 8 by both metrics
Anthropic Claude Fable 5Adjusted LLM40.800101.17Top 8 by COMET-Kiwi
Google Gemini 3.1 Pro PreviewAdjusted LLM50.79951.14Top 8 by both metrics
Anthropic Claude Opus 4.8Adjusted LLM60.798141.22Top 8 by COMET-Kiwi
Cohere Command A TranslateAdjusted LLM70.79831.12Top 8 by both metrics
Anthropic Claude Sonnet 4.6Adjusted LLM80.79761.14Top 8 by both metrics
OpenAI GPT-5.6 LunaAdjusted LLM90.79791.17Not shortlisted
OpenAI GPT-5.5Adjusted LLM100.794161.26Not shortlisted
Google Gemma 4Adjusted LLM110.79471.16Top 8 by MetricX
Mistral Large 3Adjusted LLM120.79421.11Top 8 by MetricX
OpenAI GPT-4.1Adjusted LLM130.794111.19Not shortlisted
MoonshotAI Kimi K2.5Adjusted LLM140.793131.22Not shortlisted
z.ai GLM-5Adjusted LLM150.792121.19Not shortlisted
DeepSeek V3.2Adjusted LLM160.791171.27Not shortlisted
xAI Grok 4.20 (non-reasoning)Adjusted LLM170.789191.31Not shortlisted
Mistral Medium 3.1Adjusted LLM180.788151.25Not shortlisted
Translated Lara 2Adjusted NMT190.787201.32Real-time place
Google Translation LLMAdjusted NMT200.787291.64Real-time place
Swiss AI Apertus-8B-2509Off-the-shelf LLM210.783281.53Not shortlisted
Meta Llama 4 MaverickAdjusted LLM220.782181.28Not shortlisted
Microsoft TranslatorOff-the-shelf NMT230.780221.35Not shortlisted
Anthropic Claude Haiku 4.5Adjusted LLM240.779251.38Not shortlisted
MiniMax M2.5Adjusted LLM250.778231.36Not shortlisted
Alibaba Qwen3-235B-A22BAdjusted LLM260.777241.38Not shortlisted
ModernMT Enterprise EditionOff-the-shelf NMT270.776211.33Not shortlisted
OpenAI gpt-oss-120bOff-the-shelf LLM280.776261.41Not shortlisted
Google Cloud Advanced Translation APIOff-the-shelf NMT290.770342.08Not shortlisted
Amazon TranslateOff-the-shelf NMT300.769271.48Not shortlisted
SYSTRAN PNMT APIOff-the-shelf NMT310.753301.66Not shortlisted
Tencent Cloud TMTOff-the-shelf NMT320.748311.80Not shortlisted
Yandex Cloud Translate v2Off-the-shelf NMT330.738321.93Not shortlisted
Naver Papago NMT CommercialOff-the-shelf NMT340.733331.99Not shortlisted
Alibaba Machine Translation APIOff-the-shelf NMT350.717352.25Not shortlisted
AI21 Jamba Large 1.7Adjusted LLM360.707363.19Not shortlisted
Cohere Tiny AyaAdjusted LLM370.694373.51Not shortlisted
AppTek NMTOff-the-shelf NMT380.650383.79Not shortlisted
OpenAI GPT-5.6 SolAdjusted LLM—0.795—1.15Same model family as OpenAI GPT-5.6 Luna
OpenAI GPT-5.6 TerraAdjusted LLM—0.796—1.27Same model family as OpenAI GPT-5.6 Luna
xAI Grok 4.20 (reasoning)Adjusted LLM—0.789—1.25Same model family as xAI Grok 4.20 (non-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 38 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

On MetricX, Anthropic Claude Opus 4.8 and Anthropic Claude Fable 5, the models of the two solutions that meet every requirement, rank 14 and 10. Mistral Large 3 and Google Gemma 4 made the shortlist through MetricX alone, at ranks 2 and 7. Their translations have 46.1 and 35.3 weighted errors per 1,000 words. Only the shortlisted solution on the LLM-based MT engine Google Translation LLM has more than 46.1, with 69.9 weighted errors per 1,000 words. The model nearest the cut is OpenAI GPT-5.6 Luna, 9th on both COMET-Kiwi and MetricX.

How we measure it: 00-08 · 00-09

English to Vietnamese translation cost

Quality levelCheapest solution that meets every requirement of the levelCost per million characters
BasicAnthropic Claude Opus 4.8, one-step LLM adjusted for Standard requirements$16.82
StandardAnthropic Claude Opus 4.8, one-step LLM adjusted for Standard requirements$16.82
AdvancedAnthropic Claude Opus 4.8, one-step LLM adjusted for Standard requirements$16.82
Chart data
Cost of the shortlisted one-step solutions in Vietnamese
ModelUS dollars per million charactersClass
Google Gemma 4$0.13Open-weight LLMs
Mistral Large 3$1.09LLMs
Google Gemini 3.5 Flash$3.25LLMs
Google Gemini 3.1 Pro Preview$4.37LLMs
Anthropic Claude Sonnet 5$6.68LLMs
Anthropic Claude Sonnet 4.6$8.67LLMs
Anthropic Claude Opus 4.8$16.82LLMs
Google Translation LLM$21.37LLM-based MT engines
Translated Lara 2$24.99LLM-based MT engines
DeepL Next Gen$27.5LLM-based MT engines
Anthropic Claude Fable 5$33.56LLMs
Chart data
Tokens per character of the translation into Vietnamese, by model
ModelTokens per characterClass
Mistral Medium 3.10.588LLMs
Mistral Large 30.58LLMs
AI21 Jamba Large 1.70.553LLMs
Anthropic Claude Opus 4.80.543LLMs
Anthropic Claude Fable 50.539LLMs
Anthropic Claude Sonnet 50.538LLMs
MoonshotAI Kimi K2.50.488LLMs
Anthropic Claude Haiku 4.50.483LLMs
Anthropic Claude Sonnet 4.60.483LLMs
MiniMax M2.50.405LLMs
DeepSeek V3.20.398LLMs
Alibaba Qwen3-235B-A22B0.294Open-weight LLMs
OpenAI GPT-5.6 Terra0.293LLMs
OpenAI GPT-4.10.292LLMs
OpenAI GPT-5.50.292LLMs
OpenAI GPT-5.6 Sol0.292LLMs
z.ai GLM-50.291LLMs
OpenAI GPT-5.6 Luna0.291LLMs
Cohere Command A Translate0.289LLMs
Google Gemini 3.1 Pro Preview0.286LLMs
Google Gemini 3.5 Flash0.286LLMs
Google Gemma 40.286Open-weight LLMs
xAI Grok 4.20 (non-reasoning)0.28LLMs
xAI Grok 4.20 (reasoning)0.28LLMs
Cohere Tiny Aya0.268LLMs
Cost driver

The tokenizers produce a median 0.29 tokens per character of Vietnamese, 1.3 times as many as for English. The tokenizers of Anthropic Claude Fable 5 and Anthropic Claude Opus 4.8, the models of the two solutions that meet every Standard requirement, produce 0.539 and 0.543 tokens per character, against 0.286 for Google Gemini 3.1 Pro Preview, Google Gemini 3.5 Flash and Google Gemma 4. The cost does not change from one level to the next, because the one-step solution on Anthropic Claude Opus 4.8 is the cheapest that meets each level.

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

English to Vietnamese translation speed

Quality levelFastest solution that meets every requirement of the levelSeconds per 1,000 words
BasicAnthropic Claude Opus 4.8, one-step LLM adjusted for Standard requirements67
StandardAnthropic Claude Opus 4.8, one-step LLM adjusted for Standard requirements67
AdvancedAnthropic Claude Opus 4.8, one-step LLM adjusted for Standard requirements67
Chart data
Time to translate 1,000 words into Vietnamese, by model
ModelSeconds per 1,000 wordsClass
Google Cloud Advanced Translation API0.1MT engines
Microsoft Translator0.3MT engines
ModernMT Enterprise Edition0.4MT engines
DeepL Next Gen0.5LLM-based MT engines
Google Translation LLM0.5LLM-based MT engines
Amazon Translate1.1MT engines
Translated Lara 21.3LLM-based MT engines
OpenAI gpt-oss-120b1.6Open-weight LLMs
Yandex Cloud Translate v22.3MT engines
SYSTRAN PNMT API3.5MT engines
OpenAI GPT-5.6 Luna14LLMs
Cohere Tiny Aya16LLMs
OpenAI GPT-5.6 Terra17LLMs
OpenAI GPT-4.118LLMs
xAI Grok 4.20 (non-reasoning)21LLMs
Alibaba Machine Translation API21MT engines
DeepSeek V3.226LLMs
Cohere Command A Translate30LLMs
OpenAI GPT-5.6 Sol33LLMs
Anthropic Claude Haiku 4.538LLMs
Swiss AI Apertus-8B-250940Open-weight LLMs
Google Gemini 3.5 Flash43LLMs
MiniMax M2.544LLMs
OpenAI GPT-5.547LLMs
Mistral Medium 3.147LLMs
Meta Llama 4 Maverick50Open-weight LLMs
Anthropic Claude Sonnet 551LLMs
Mistral Large 351LLMs
Tencent Cloud TMT59MT engines
Google Gemini 3.1 Pro Preview65LLMs
Anthropic Claude Opus 4.867LLMs
Anthropic Claude Fable 568LLMs
Naver Papago NMT Commercial68MT engines
xAI Grok 4.20 (reasoning)68LLMs
Anthropic Claude Sonnet 4.675LLMs
AI21 Jamba Large 1.775LLMs
Alibaba Qwen3-235B-A22B93Open-weight LLMs
AppTek NMT101MT engines
Google Gemma 4134Open-weight LLMs
z.ai GLM-5140LLMs
MoonshotAI Kimi K2.5185LLMs

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 Vietnamese.

Real-time

Five of the nine real-time models in Vietnamese are MT engines, three are LLM-based MT engines and one is the open-weight OpenAI gpt-oss-120b. The three real-time solutions on the shortlist miss general translation quality and formatting. Among the LLMs, open-weight ones included, translation times range from 1.6 seconds per 1,000 words, with OpenAI gpt-oss-120b, to 185, with MoonshotAI Kimi K2.5. The time does not change from one level to the next: the fastest solution that meets each level takes 67 seconds per 1,000 words.

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

Tone of voice and gender-neutral English to Vietnamese translation

Tone of voice
Formal register, with quý vị or neutral terms
Gender-neutrality strategy
neutral role nouns (người-) + first-person plural (chúng tôi) + neutral noun or rephrasing
Options in the language
gendered kinship pronouns (anh/chị/ông/bà), gendered titles, neutral 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
Anthropic Claude Opus 4.8one-step100%✓$16.8267
Anthropic Claude Fable 5one-step100%✓$33.5668
Analysis

For tone of voice in Vietnamese we check the formal register, with quý vị or neutral terms. In translations from DeepL Next Gen, Cohere Command A Translate and Google Translation LLM, 82.98%, 78.7% and 68.1% of the sentences we check are in the formal register. For gender neutrality we chose neutral role nouns with người, the first-person plural chúng tôi, and neutral nouns or rephrasing. Gendered kinship pronouns (anh, chị, ông, bà) and gendered titles are the other options, and a default gender is never allowed.

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

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

Confidence

The two metric pairs agree above the threshold in Vietnamese, and neither is borderline. The three LQA judges agree on a median of 95.2% of the sentences of the shortlisted solutions. So both the shortlist, which we build on the metrics, and the Intento LQA results on this page rest on agreeing measurements.

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