Machine Translation Units (MTUs) are the metric Phrase uses to measure the consumption of machine translation services accessed through Phrase Language AI and/or Phrase Portal.
Monitoring consumption
Monitor machine translation (MT) usage in the tab of Phrase Language AI.
The tab presents:
-
Machine Translation Units (MTUs) and translated characters consumed by project, workflow step, locale pair, MT engine, and MT profile.
-
Comparison of translated content vs. MTU consumption.
-
The TMS MT origin filter shows the split of MT units across Phrase TMS:
-
TMS pre-translation
-
TMS analysis
-
TMS CAT editor
MT units consumed prior to the 1st of December, 2024 are not split by origin and are marked as Uncategorized.
-
MTU Consumption
Phrase calculates Machine Translation Units (MTUs) based on the number of input characters sent to the Phrase Language AI service.
The number of input characters is calculated based on the number of Unicode code units within a string (unlike how calculations are made for analysis).
Note
The number of input characters does not directly correspond to the size of an uploaded document or file.
Examples:
-
A file with very good translation memory (TM) matches would have a far smaller number of input characters for MTU than a file with little to no TM matches.
-
If a single file is processed by MT multiple times (once in pre-translation, once in analysis and then again in analysis during a different workflow step), this would triple the number of input characters.
The charging rates for MTUs depend on the type of engine configured:
-
Native Engines
1 input character = 1 MTU -
Ecosystem Engines
1 input character = 0.5 MTU-
A fixed fee of 0.5 MTUs per input character for accessing Phrase Language AI (PLAI) services is calculated regardless how many benefits are used.
-
Benefits of PLAI:
-
Centralized MT profile configuration with custom translation settings and integration across the Phrase Suite.
-
Access to Phrase Quality Performance Score (QPS) for translation quality assessments.
-
Use of custom translation portals for organizational MT needs.
-
Access to Next MT and Next GenMT (LLM-based) engines.
-
Support for custom-trained engines via Custom AI.
-
-
-
Bring Your Own Engines (BYO Engines)
1 input character = 0.5 MTU
Consumption management
The number of MTUs consumed depends on configuration choices and content type. In Phrase TMS, adjusting the settings for pre-translation, analysis and the CAT editor can optimize MTU consumption.
-
A TM threshold sets when pre-translation uses MT and as MT has lower editing times, higher TM thresholds reduce human post-editing efforts. Pre-translation only uses MT if no TM match is found at or above the set threshold. Using the recommended threshold is likely to reduce total translation costs.
-
In the global settings or for a specific project or project template, ensure is enabled.
Example scenarios:
-
Documentation with strong TM leverage (e.g. technical documentation, structured or repetitive text)
MTU usage is minimized to rely on TM matches. Pre-translating with a lower TM threshold is recommended.
-
Marketing or creative content (e.g. campaign copy, product descriptions, website text)
MTU usage is acceptable to reduce post-editing effort and improve fluency. Pre-translating with a higher TM threshold is recommended.
Tip
TM penalties can lower matches below the set threshold, which may cause MT to be called unnecessarily. Adjust penalties and thresholds together to control MTU usage.
Higher thresholds equals more MT usage which equals more MTUs (because fewer segments are covered by TM and MT is triggered more often).
Lower thresholds equal less MT usage which equal fewer MTUs (more segments are satisfied by TM and MT is skipped).
MT enabled in Pre-translation
Apply MT only when there’s no TM ≥ threshold
Result
MTU consumption
Off
Off
Pre-translation never calls MT.
No MTUs consumed.
Off
On
Pre-translation never calls MT.
No MTUs consumed.
On
Off
MT can be applied regardless of TM matches (no threshold check).
MTUs consumed for segments where MT is applied
On
On
MT is applied only when there’s no TM match ≥ threshold.
Reduced MTUs: only segments below the threshold call MT
-
-
The standard analysis is run on uploaded source segments before translation to assess translation effort and serve as the baseline for generating quotes for clients.
MT is only consumed if a quality performance score (QPS) is required and if there are no 100/101% TM or non-translatables (NT) matches.
If not required:
-
Disable the option.
-
Ensure 100/101% TM or non-translatables (NT) matches are prioritized over MT matches in pre-translation settings.
Note
Post-editing analysis does not consume MTUs.
Turning QPS off is the only way to make sure analysis never consumes MTUs.
Keeping QPS on but combining it with reduces unnecessary MT queries if MT leverage needs to be seen in analysis.
QPS (Include MT matches)
Apply MT only when no TM ≥ threshold
MTU consumption in Analysis
Off
Off
No MTUs consumed (MT not queried at all)
Off
On
No MTUs consumed (QPS is off → Apply MT setting has no effect in Analysis)
On
Off
MTUs consumed for all segments without 100/101% TM/NT matches (even if fuzzy TM matches exist, MT still queried)
On
On
Reduced MTU consumption → MTUs consumed only for segments with no TM ≥ threshold; segments with 100/101% or other high TM matches are excluded from MT queries
-
-
-
In the settings for the project or project template, enable .
-
Enabling this option controls when MT is suggested in the CAT pane. MT will only be used for the segments with a TM match below the set threshold.
-
If segments are confirmed (and thus saved to TM) in a workflow step, they will have a 101% TM match in subsequent steps. As a result, MT will not be suggested, preserving MTU in those steps.
-
If this option is disabled, MT is suggested regardless of the threshold, leading to MTU consumption.
-
-
A further optimization of MTU consumption can be achieved by pre-translating a job before opening it in the editor.
If MT is used as the target during pre-translation, the CAT pane reuses the MT suggestion without making a new MT request. The reused suggestion is carried through all workflow steps without calling MT again.
This optimization is enabled by default (without any settings control).
Note
This optimization is not available in the CAT desktop editor.
MTUs are still consumed in the following cases:
-
No pre-translation is performed on the job.
-
MT is not used for a segment during pre-translation (e.g., NT is used instead).
-
A TM match is used in pre-translation, and MT is called if the TM match does not meet the threshold defined for the CAT editor.
-
The source text is changed, or a segment is joined/split.
-
The MT engine is changed after pre-translation.
-
The source text is automatically updated during a source update or as part of a continuous job.
Higher threshold - more MT usage - more MTUs (because fewer segments are covered by TM and thus MT is triggered more often).
Lower threshold - less MT usage - fewer MTUs (more segments are satisfied by TM and skip MT).
-
-
MTUs are consumed both when working on segments within the editor and when segments are pre-loaded:
-
CAT results are pre-loaded to minimize wait times, consuming MTU for the previous segment, the current segment, and the next two segments.
-
When navigating with keyboard arrows or using , MTU are consumed for these pre-loaded segments.
-
When jumping upon confirmation, MTU are consumed for the next two segments based on the in the editor.
-
Pre-loading respects filtering results, ensuring that MTU are consumed only for segments displayed after filtering.
-
-
MTUs are not consumed repeatedly for revisited segments within the same user session. A user session remains active as long as the job is open in the browser.
-
By default, the maximum editor user session lasts three days (unless a shorter session duration is set in TMS). After this period, the user is required to refresh the job.
-
Refreshing the job, whether due to session expiration or manual refresh, ends the current user session.
-
Upon refreshing, a new user session begins, and MTUs are consumed again for the relevant segments.
-
-
User role rights
Threshold ON ensures MT is only called if TM is too weak (< threshold).
Pre-translation with MT → saves MTUs, as the editor reuses them (web editor only).
Pre-translation with strong TM (≥ threshold) → no MTUs used.
Pre-translation with weak TM (< threshold) → MT still queried in CAT, consuming MTUs.
Suggest MT only for segments below threshold
Pre-translation situation
CAT editor suggestions (CAT pane)
MTU consumption
Off
No pre-translation
MT is suggested for all segments, regardless of TM matches.
MTUs consumed
Off
Segment pre-translated with MT
CAT reuses MT suggestions from pre-translation, no new MT calls.
No MTUs consumed
On
No pre-translation
MT is suggested only for segments with TM below threshold.
Reduced MTU consumption
On
Segment pre-translated with MT
CAT reuses MT suggestions from pre-translation, no new MT calls.
No MTUs consumed
On
Segment pre-translated with TM ≥ threshold
MT not suggested (TM covers it).
No MTUs consumed
On
Segment pre-translated with TM ≥ threshold
MT is still called in the editor, because TM didn’t meet the threshold.
MTUs consumed
-
-
MT is always used, and third-party engines outside PLAI cannot be configured.
-
To reduce MTU consumption, use an ecosystem engine with credentials to consume only 0.5 MTU per character.
-
-
When Phrase Language AI is selected as machine translation provider, MTUs from the Phrase Platform subscription allowance are always consumed.
-
To reduce MTU consumption, use any other 3rd party MT engine available in Phrase Strings.
-
-
When Phrase Language AI is selected as translation AI model, MTUs from the Phrase Platform subscription allowance are always consumed.
-
To reduce MTU consumption, use any other translation AI model available in Phrase Studio.
-