Part of Best LLM for Chat & Roleplay
Best LLM for Roleplay & Character
Gemini 3.8 Flash ranks first for Roleplay & Character as of October 20, 2026.[1][2] It scores 81.41% on LiveBench Instruction Following (#1 of 59), including paraphrasing, simplifying, story generation, and summarization. Gemini 3.7 Flash is the next-ranked alternative. It scores 79.93% on LiveBench Instruction Following (#2 of 59), including paraphrasing, simplifying, story generation, and summarization.
- Best non-Google pick: Claude Fable 5.1 (#4)
- Compare Gemini 3.8 Flash vs Gemini 3.7 Flash →
About this recommendation
- Updated
- Evidence through
- Sources
- 11
- Revision
- v102
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
100%
intended feed weight
Largest provider share
5 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 |
|---|---|---|---|
| LMArena Creative WritingObserved 2026-10-15 · evaluation date not supplied · high effort | 39% | #6 | 24/24 |
| LMArena TextObserved 2026-10-15 · evaluation date not supplied · high effort | 28% | #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 |
| LiveBench Language (2026-06-25)Observed 2026-10-18 · evaluation date not supplied · gemini-3.8-flash-highfailing; last successful check 2026-10-18 | 11% | #6 | 24/24 |
Provider concentration
Each exact model is scored separately; provider identity is not a ranking input.
- Google5 models
- Anthropic2 models
- 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 100%
#1 LiveBench Instruction Following · #6 LiveBench Language
Google · coverage 100%
#2 LiveBench Instruction Following · #8 LMArena Creative Writing
Google · coverage 100%
#3 LiveBench Instruction Following · #11 LMArena Creative Writing
Anthropic · coverage 100%
#3 LiveBench Language · #3 LMArena Creative Writing
Meta · coverage 100%
#4 LiveBench Instruction Following · #9 LMArena Text
Anthropic · coverage 100%
#1 LMArena Creative Writing · #3 LMArena Text
Google · coverage 100%
#6 LiveBench Instruction Following · #12 LMArena Creative Writing
Google · coverage 100%
#8 LiveBench Instruction Following · #14 LMArena Creative Writing
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
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.
- 03
Gemini 3.1 Pro PreviewGoogle · $2 in / $12 out per M tokens · 1.0M context
Gemini 3.1 Pro Preview scores 79.1% on LiveBench Instruction Following (#3 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
Claude Fable 5.1Anthropic · $10 in / $50 out per M tokens · 1M context
Ranks #3 on LMArena creative-writing with Elo 1508 and #3 on LiveBench Language, demonstrating elite human-preferred creative prose and strong language manipulation.
Best when: Use for character voice consistency and creative prose where human voters ranked it 3rd of 152 models, with an Elo of 1508.More tips
- Deploy for language manipulation tasks like tone shifting and stylistic adaptation where it scored 89.5%, 3rd in the catalog.
Watch out for
- Instruction following drops to #14 of 59 on LiveBench, so verify it captures complex narrative constraints in your specific roleplay setup.
- 05
Muse Spark 1.3Meta · $1.25 in / $4.25 out per M tokens · 1.0M context
Muse Spark 1.3 scores 78% on LiveBench Instruction Following (#4 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.
- 06
Claude Opus 5.5Anthropic · $4 in / $20 out per M tokens · 1M context
Claude Opus 5.5 ranks #1 of 152 on LMArena's creative-writing category (Elo 1534), based on blind human preference votes.
Best when: Its strongest result for this task is on LMArena Creative Writing, which measures creative writing people prefer in blind comparison.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 07
Gemini 3.5 FlashGoogle · $1.50 in / $9 out per M tokens · 1.0M context
Ranks #6 on LiveBench Instruction Following and #12 on LMArena creative-writing, balancing reliable narrative direction adherence with moderate creative preference.
Best when: Use for simplifying and paraphrasing character voices where it placed 6th in objective instruction-following evaluation.Watch out for
- Overall text arena ranking drops to #23 of 152, indicating broader context handling may lag behind specialized creative performance.
- 08
Gemini 3.6 FlashGoogle · $0.75 in / $3.75 out per M tokens · 1.0M context
Ranks #8 on LiveBench Instruction Following and #14 on LMArena creative-writing, showing competent but not exceptional performance across creative benchmarks.
Best when: Use for story generation and summarization where it maintained 75.37% on objective instruction-following evaluation.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
Frequently asked
- What is the best model for Roleplay & Character?
- Gemini 3.8 Flash ranks first as of October 20, 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?
- Gemini 3.7 Flash ranks second. Gemini 3.7 Flash scores 79.93% on LiveBench Instruction Following (#2 of 59), including paraphrasing, simplifying, story generation, and summarization.[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 79.93% on LiveBench Instruction Following (#2 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026 - 3
“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 - 4
“Within the 152 catalog models matched to this evaluation: Ranks #3 of 152 on LMArena's creative-writing category (Elo 1508), based on blind human preference votes.”
LMArena creative-writing category · Benchmark · Oct 15, 2026 - 5
“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 - 6
“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 - 7
“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 - 8
“Within the 152 catalog models matched to this evaluation: Ranks #1 of 152 on LMArena's creative-writing category (Elo 1534), based on blind human preference votes.”
LMArena creative-writing category · Benchmark · Oct 15, 2026 - 9
“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 - 10
“Within the 152 catalog models matched to this evaluation: Ranks #23 of 152 on LMArena's overall text arena (Elo 1477), based on blind human preference votes.”
LMArena text arena · Benchmark · Oct 15, 2026 - 11
“Within the 59 catalog models matched to this evaluation: Scores 75.37% on LiveBench Instruction Following (#8 of 59), including paraphrasing, simplifying, story generation, and summarization.”
LiveBench Instruction Following · Benchmark · Oct 18, 2026
Rankings synthesized from community evidence and open benchmarks. See methodology. Not driven by vendor marketing.