Llama-3.1-Tulu-3-70B

Ai2
Arena score
1256
Among open models
#101of 148
Among all on Arena
#297
Votes
2,846

Measured on Llama 3.1 Tulu 3 70B, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.

Score by category

CategoryScore
  • English1273
  • Science1267
  • Writing & literature1254
  • Multi-turn1251
  • Chinese1249
  • Applied math1246
  • Russian1245
  • Math1242
  • Software & IT1242
  • Media & sports1240
  • Hard prompts, English1236
  • Non-English1236
  • Coding1235
  • Instructions1232
  • Creative writing1231
  • Business & finance1226
  • Long prompts1224
  • Hard prompts1220
  • Excluding ties1170

Score by size

Plot against
Best for its size
ModelParamsScore
Llama-3.1-Tulu-3-8B8.0B1193
Llama-3.1-Tulu-3-70B70.6B1256

About as strong, smaller

Compare

Built on it

Quantizations
10
Fine-tunes
0
Adapters
1
Merges
12

Ways to run it

Most downloaded

BenchClub runs

Not runnable yet

Runs open once a Llama-3.1-Tulu-3-70B build is registered.

More from Ai2

See more

Other 40–80B models

See more