Gemini 4 Argon
Claude Opus 5.5
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 Preview
Llama 3.3 70B Instruct
Llama 3.3 70B Instruct is an open-weight model from Meta, released Dec 6, 2024. It ranks #278 of 351 on the LMArena text leaderboard with a rating of 1274. It costs $0.10 per million input tokens and $0.32 per million output tokens, with a 128K-token context window.
- Developer
- Meta
- Released
- Dec 6, 2024
- Context window
- 128K
- Price in / out, per 1M tokens
- $0.10 / $0.32
- Weights
- Open · Llama-3.3
- Inputs
- Text
- Reasoning
- No
- Tool use
- Yes
- Hugging Face downloads, 30 days
- 251.9K
- Hugging Face likes
- 3.2K
The rating LMArena published at each update since Dec 15, 2024. Ratings move as votes come in and new models join.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1274 | 278 / 351 | 54.5K |
| CodingLMArena | 1268 | 286 / 346 | 8.7K |
| MathLMArena | 1267 | 273 / 339 | 5.8K |
| Creative writingLMArena | 1250 | 268 / 349 | 8K |
| Instruction followingLMArena | 1242 | 286 / 351 | 18.8K |
| Hard promptsLMArena | 1257 | 280 / 351 | 17K |
| Multi-turnLMArena | 1280 | 265 / 349 | 9.1K |
| Long queriesLMArena | 1255 | 285 / 330 | 8.3K |
| ExpertLMArena | 1225 | 275 / 305 | 2.9K |
| Legal and governmentIndustries | 1284 | 266 / 325 | 3.2K |
| Medicine and healthcareIndustries | 1275 | 263 / 325 | 2.7K |
| Business and financeIndustries | 1243 | 280 / 344 | 6.6K |
| ChineseLanguages | 1217 | 300 / 328 | 3.7K |
| SpanishLanguages | 1270 | 233 / 249 | 636 |
| GermanLanguages | 1251 | 232 / 261 | 1.2K |
| FrenchLanguages | 1282 | 228 / 252 | 575 |
| JapaneseLanguages | 1150 | 234 / 236 | 1.2K |
LMArena, CC BY 4.0, as of Oct 22, 2026. Where a model is rated at several effort levels, its best is shown.
Compare Llama 3.3 70B Instruct
Meta
Meta’s AI research publishes the Llama family, among the most downloaded open-weight models. Founded 2013 (FAIR), based in Menlo Park. ai.meta.com
Questions
Who makes Llama 3.3 70B Instruct?
Llama 3.3 70B Instruct is made by Meta, founded 2013 (FAIR) and based in Menlo Park.
How much does Llama 3.3 70B Instruct cost?
$0.10 per million input tokens and $0.32 per million output tokens, at its provider’s list price.
What is Llama 3.3 70B Instruct’s context window?
128K tokens (131,072).
How does Llama 3.3 70B Instruct rank?
#278 of 351 on LMArena’s text leaderboard, rated 1274 (95% interval 1271–1278) from 54,493 votes, as of Oct 22, 2026.
Is Llama 3.3 70B Instruct open source?
Llama 3.3 70B Instruct has open weights under the Llama-3.3 license, downloadable from Hugging Face (meta-llama/Llama-3.3-70B-Instruct).