Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
GPT-5.5
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 PreviewClaude Opus 4.6 vs Gemini 3.8 Flash
- Claude Opus 4.6 and Gemini 3.8 Flash are level on LMArena’s text leaderboard (1502 and 1498).
- Gemini 3.8 Flash is cheaper: $0.82 / $4.13 per million input / output tokens, against $5 / $25 for Claude Opus 4.6.
- On the 7 Vals AI benchmarks both have taken, Claude Opus 4.6 wins 1 and Gemini 3.8 Flash wins 6 (Claude Opus 4.6: MedCode).
LatestClaude Opus 4.6 1502
Side by side
Models (2/6)
- Claude Opus 4.6
- Gemini 3.8 Flash
| Benchmark | Claude Opus 4.6 | Gemini 3.8 Flash |
|---|---|---|
| The model | ||
| Lab | Anthropic | |
| Released | 2026-02-04 | 2026-09-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Input, per 1M tokens | $5 | $0.82 |
| Output, per 1M tokens | $25 | $4.13 |
| LMArena, people’s votes | ||
| LMArena TextMillions of people compare two anonymous models side by side and vote for the better answer | 1502#6 of 351 | 1498#8 of 351 |
| LMArena WebDevPeople ask two anonymous models to build the same web app, try both and vote for the better one | 1546#45 of 124 | 1585#36 of 124 |
| LMArena VisionSide-by-side votes on prompts that include an image | 1316#6 of 146 | 1315#7 of 146 |
| LMArena CodingLMArena’s text votes on prompts that ask for code | 1534#3 of 346 | 1508#17 of 346 |
| LMArena MathLMArena’s text votes on math prompts | 1518#7 of 339 | 1532#4 of 339 |
| LMArena Creative writingLMArena’s text votes on stories, poems and other creative writing | 1505#4 of 349 | 1489#8 of 349 |
| LMArena Hard promptsLMArena’s text votes on the hardest prompts | 1524#5 of 351 | 1509#9 of 351 |
| LMArena Instruction followingLMArena’s text votes on prompts with specific instructions to follow | 1521#3 of 351 | 1492#11 of 351 |
| Vals AI, overall | ||
| Vals IndexVals AI’s headline score | — | 55.95%#15 of 27 |
| Vals AI, coding | ||
| Terminal-Bench 4.0Frontier-difficulty tasks an agent completes in a real terminal | — | 19.70%#31 of 55 |
| SWE-benchReal GitHub issues from open-source projects | 78.20%#30 of 85 | 80.00%#23 of 85 |
| Vibe Code BenchBuilding complete web apps from a plain-language brief, graded on whether the app works | 53.50%#57 of 115 | 78.65%#29 of 115 |
| Code MigrationRewriting working programs in a new language, graded on whether the new program still passes its tests | — | 38.65%#22 of 28 |
| LiveCodeBenchFresh competitive programming problems, collected after models’ training cut-offs to avoid contamination | 84.68%#38 of 143 | 89.48%#3 of 143 |
| Vals AI, finance | ||
| Finance AgentThe core tasks of a financial analyst, done as an agent with research tools | — | 65.76%#5 of 31 |
| Finance IndexVals AI’s composite of finance tasks | — | 60.46%#9 of 31 |
| Tax Agent BenchAgentic research into US corporate tax questions | — | 32.12%#21 of 76 |
| Vals AI, law | ||
| Legal IndexVals AI’s composite of legal tasks | — | 69.78%#14 of 28 |
| Legal Research BenchAgentic US legal research | — | 38.94%#34 of 85 |
| Vals AI, medicine | ||
| MedCodeAssigning the right medical billing codes from clinical notes | 48.24%#30 of 107 | 48.13%#31 of 107 |
| MedQAMedical licensing exam questions, measured on accuracy and on robustness to bias | 95.41%#12 of 92 | — |
| Vals AI, reasoning | ||
| GPQA DiamondGraduate-level questions in biology, physics and chemistry that experts answer well and search does not | 89.65%#29 of 138 | 94.44%#4 of 138 |
| MMLU-ProHarder, reasoning-focused multiple choice across 14 subjects, from law to engineering | 89.11%#15 of 138 | 90.22%#5 of 138 |
| AIMEProblems from the American Invitational Mathematics Examination, a qualifier for the US Math Olympiad | 95.63%#9 of 96 | — |
| MMMUCollege-level questions that need both an image and text | 83.87%#29 of 93 | 89.08%#4 of 93 |
| Terminal-Bench ScienceExpert-written scientific research workflows, run by an agent in a terminal | — | 8.57%#19 of 42 |
Add models in the compare tool.
Questions
Which is better, Claude Opus 4.6 or Gemini 3.8 Flash?
Claude Opus 4.6 and Gemini 3.8 Flash are level on LMArena’s text leaderboard (1502 and 1498). Gemini 3.8 Flash is cheaper: $0.82 / $4.13 per million input / output tokens, against $5 / $25 for Claude Opus 4.6. On the 7 Vals AI benchmarks both have taken, Claude Opus 4.6 wins 1 and Gemini 3.8 Flash wins 6 (Claude Opus 4.6: MedCode).
Which is cheaper, Claude Opus 4.6 or Gemini 3.8 Flash?
Gemini 3.8 Flash is cheaper: $0.82 / $4.13 per million input / output tokens, against $5 / $25 for Claude Opus 4.6.