Part of Best LLM for Translation
Best LLM for Literary Translation
Gemini 3.8 Flash ranks first for Literary Translation as of October 21, 2026.[1][2] It scores 81.41% on LiveBench Instruction Following (#1 of 59), including paraphrasing, simplifying, story generation, and summarization. Claude Fable 5.1 is the next-ranked alternative. It achieves #3 on LiveBench Language with 89.5%, indicating strong performance on objective language manipulation tasks.
- Best non-Google pick: Claude Fable 5.1 (#2)
- Compare Gemini 3.8 Flash vs Claude Fable 5.1 →
About this recommendation
- Updated
- Evidence through
- Sources
- 15
- Revision
- v106
Decision audit
Why this result
Inspect the inputs and the computed order behind the recommendation.
Models screened
24
live candidates
Evaluation feeds
4
task-weighted
Winner coverage
50%
intended feed weight
Largest provider share
4 of 8
Google: Gemini 3.8 Flash leads the measured task comparison with 2 independent evaluation families. Community volume, release age and usage popularity add no ranking points.
Real-world track record is still developing. The recommendation can lead on independent measurements while practitioner reports accumulate.
Sources evaluated
These are the weights used in scoring. Missing measurements are omitted from each model’s average. A feed with no result for any model released in the last 60 days is stale and does not score until it catches up.
| Evaluation feed | Weight | Winner result | Field measured |
|---|---|---|---|
| WMT translationdisabled; last successful check unknownunavailable | 0% | feed unavailable | 0/24 |
| LiveBench Language (2026-06-25)Observed 2026-10-18 · evaluation date not supplied · gemini-3.8-flash-highfailing; last successful check 2026-10-18 | 44% | #6 | 24/24 |
| LMArena TextObserved 2026-10-20 · evaluation date not supplied · high effort | 33% | #8 | 24/24 |
| LiveBench Instruction Following (2026-06-25)Observed 2026-10-18 · evaluation date not supplied · gemini-3.8-flash-highfailing; last successful check 2026-10-18 | 22% | #1 | 24/24 |
Provider concentration
Each exact model is scored separately; provider identity is not a ranking input.
- Google4 models
- OpenAI2 models
- Anthropic1 model
- Meta1 model
Decision table
Every published model is shown in computed order. Practitioner sources are distinct community threads, not the citations repeated in the prose below.
Google · coverage 50%
#1 LiveBench Instruction Following · #6 LiveBench Language
Anthropic · coverage 50%
#3 LiveBench Language · #5 LMArena Text
Google · coverage 50%
#2 LiveBench Instruction Following · #11 LiveBench Language
OpenAI · coverage 50%
#4 LiveBench Language · #7 LiveBench Instruction Following
OpenAI · coverage 50%
#2 LiveBench Language · #11 LiveBench Instruction Following
Google · coverage 50%
#3 LiveBench Instruction Following · #12 LiveBench Language
Meta · coverage 50%
#4 LiveBench Instruction Following · #9 LMArena Text
Google · coverage 50%
#6 LiveBench Instruction Following · #14 LiveBench Language
Task score averages the available, task-weighted measurements. Benchmark positions compare models matched to our catalog. Models measured by at least two independent evaluator families are considered first. Missing tests, practitioner report counts, popularity and release age add no points. Coverage describes evidence breadth, not a probability of correctness. Scores within two points are shown as close under this policy; this is not a statistical confidence interval. When the leading scores are that close, the model measured by more evaluator families ranks first.
- 01
Gemini 3.8 FlashGoogle · $0.83 in / $4.13 out per M tokens · 1.0M context
Gemini 3.8 Flash scores 81.41% on LiveBench Instruction Following (#1 of 59), including paraphrasing, simplifying, story generation, and summarization.
Best when: Its strongest result for this task is on LiveBench Instruction Following, which measures following detailed instructions exactly.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 02
Claude Fable 5.1Anthropic · $10 in / $50 out per M tokens · 1M context
Claude Fable 5.1 achieves #3 on LiveBench Language with 89.5%, indicating strong performance on objective language manipulation tasks.
Best when: Prioritize for raw language manipulation benchmarks where it scores 89.5% on LiveBench Language (#3 of 59).More tips
- Deploy when human preference rankings matter, as it holds #5 on LMArena's overall text arena with Elo 1503.
Watch out for
- Watch for weaker instruction following at #14 of 59 (72.99%) on LiveBench, which covers paraphrasing and story generation tasks.
- 03
Gemini 3.7 FlashGoogle · $0.75 in / $3.75 out per M tokens · 1.0M context
Gemini 3.7 Flash scores 79.93% on LiveBench Instruction Following (#2 of 59), including paraphrasing, simplifying, story generation, and summarization.
Best when: Its strongest result for this task is on LiveBench Instruction Following, which measures following detailed instructions exactly.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 04
GPT-6 AstraOpenAI · $10 in / $50 out per M tokens · 1.1M context
GPT-6 Astra scores 89.43% on LiveBench Language (#4 of 59), an objective evaluation of language manipulation tasks.
Best when: Its strongest result for this task is on LiveBench Language, which measures precise language understanding and editing.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 05
GPT-6.1 SolOpenAI · $2 in / $10 out per M tokens · 1.1M context
GPT-6.1 Sol achieves the #2 rank on LiveBench Language with 90.13%, the highest language manipulation score in this pool.
Best when: Prioritize for pure language manipulation benchmarks where it scores 90.13% and ranks #2 on LiveBench Language.Watch out for
- Account for weaker instruction following at #11 of 59 (74.15%) when tasks require paraphrasing or story generation.
- 06
Gemini 3.1 Pro PreviewGoogle · $2 in / $12 out per M tokens · 1.0M context
Gemini 3.1 Pro Preview ranks #3 on LiveBench Instruction Following, suggesting reliable performance on structured text generation tasks.
Best when: Its strongest result for this task is on LiveBench Instruction Following, which measures following detailed instructions exactly.Watch out for
- Note its #12 ranking on LiveBench Language (85.38%) places it below several competitors in this evaluation.
- 07
Muse Spark 1.3Meta · $1.25 in / $4.25 out per M tokens · 1.0M context
Muse Spark 1.3 ranks #4 on LiveBench Instruction Following, indicating strong capability for paraphrasing and text restructuring tasks.
Best when: Deploy for instruction-heavy workflows where it scores 78% and ranks #4 on LiveBench Instruction Following.Watch out for
- Expect weaker performance on language manipulation benchmarks at #20 of 59 (82.79%) on LiveBench Language.
- 08
Gemini 3.5 FlashGoogle · $1.50 in / $9 out per M tokens · 1.0M context
Gemini 3.5 Flash shows cross-script substitution defects in Norwegian compound words, with foreign-script characters replacing subword tokens.
Best when: Consider for instruction following tasks where it ranks #6 of 59 (75.6%) on LiveBench.Watch out for
- Avoid for Norwegian or similar morphologically complex languages due to verified cross-script substitution defects where single subword tokens are replaced with foreign-script characters.
- Note its #14 ranking on LiveBench Language (84.58%) falls below top performers in this evaluation.
Frequently asked
- What is the best model for Literary Translation?
- Gemini 3.8 Flash ranks first as of October 21, 2026. Gemini 3.8 Flash scores 81.41% on LiveBench Instruction Following (#1 of 59), including paraphrasing, simplifying, story generation, and summarization.[1]
- What is the best alternative to Gemini 3.8 Flash?
- Claude Fable 5.1 ranks second. Claude Fable 5.1 achieves #3 on LiveBench Language with 89.5%, indicating strong performance on objective language manipulation tasks.[2]
Sources
- 1
“Within the 59 catalog models matched to this evaluation: Scores 81.41% on LiveBench Instruction Following (#1 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 2
“Within the 59 catalog models matched to this evaluation: Scores 89.5% on LiveBench Language (#3 of 59), an objective evaluation of language manipulation tasks.”
LiveBench Language · Benchmark · Oct 18, 2026 - 3
“Within the 150 catalog models matched to this evaluation: Ranks #5 of 150 on LMArena's overall text arena (Elo 1503), based on blind human preference votes.”
LMArena text arena · Benchmark · Oct 20, 2026 - 4
“Within the 59 catalog models matched to this evaluation: Scores 72.99% on LiveBench Instruction Following (#14 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 5
“Within the 59 catalog models matched to this evaluation: Scores 79.93% on LiveBench Instruction Following (#2 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 6
“Within the 59 catalog models matched to this evaluation: Scores 89.43% on LiveBench Language (#4 of 59), an objective evaluation of language manipulation tasks.”
LiveBench Language · Benchmark · Oct 18, 2026 - 7
“Within the 59 catalog models matched to this evaluation: Scores 90.13% on LiveBench Language (#2 of 59), an objective evaluation of language manipulation tasks.”
LiveBench Language · Benchmark · Oct 18, 2026 - 8
“Within the 59 catalog models matched to this evaluation: Scores 74.15% on LiveBench Instruction Following (#11 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 9
“Within the 59 catalog models matched to this evaluation: Scores 79.1% on LiveBench Instruction Following (#3 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 10
“Within the 59 catalog models matched to this evaluation: Scores 85.38% on LiveBench Language (#12 of 59), an objective evaluation of language manipulation tasks.”
LiveBench Language · Benchmark · Oct 18, 2026 - 11
“Within the 59 catalog models matched to this evaluation: Scores 78% on LiveBench Instruction Following (#4 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 12
“Within the 59 catalog models matched to this evaluation: Scores 82.79% on LiveBench Language (#20 of 59), an objective evaluation of language manipulation tasks.”
LiveBench Language · Benchmark · Oct 18, 2026 - 13
“Within the 59 catalog models matched to this evaluation: Scores 75.6% on LiveBench Instruction Following (#6 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 14
“Note: This is a model/API-level report, filed here because #1969 tracked the same defect class (spurious foreign-script characters in output) and went stale-closed. Everything below was reproduced with raw REST calls to generativelanguage.googleapis.com/v1beta — no SDK involved — so this is not an SDK decoding artifact. Seed-based deterministic reproductions are included. Summary gemini-3.5-flash-lite and gemini-3.5-flash occasionally replace ONE subword token inside a Norw…”
Apaaaaaaaaaa · GitHub · Aug 7, 2026 - 15
“Within the 59 catalog models matched to this evaluation: Scores 84.58% on LiveBench Language (#14 of 59), an objective evaluation of language manipulation tasks.”
LiveBench Language · Benchmark · Oct 18, 2026
Rankings synthesized from community evidence and open benchmarks. See methodology. Not driven by vendor marketing.