LLMBoxing
Llama 2-70B 22679.0 People see one question with two anonymous model answers and pick the better one. The site counts preferences between model pairs.
Independent human preference site cited in the Mistral 7B paper. The domain now serves an unrelated gambling page, so only archived copies of the leaderboard remain.
- Made by
- llmboxing.com
- Released
- October 2023
- Kind
- Overall chat quality
- Level
- Everyday
- Scored
- Ranked by votes
- Answers
- Open-ended
- Strongest model tested
- Llama 2-70B #209 on the capability index
- Last in a system card
- October 2023
- Last model added
- October 2023
- Scores refreshed
- November 12th, 2026
Llama 2-70B leads with 22679.0
Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 1 Llama 2-70B Llama 2 70B vs GPT-3.5; archive 2023-12-25 22679.0
- 2 Mistral 7B Instruct [snapshot unspecified] Llama 2 13B vs Mistral 7B; archive 2023-12-25 16615.0
- 3 Llama 2-13B Llama 2 13B vs Mistral 7B; archive 2023-12-25 14029.0
- 4 GPT 3.5 [snapshot unspecified] Llama 2 70B vs GPT-3.5; archive 2023-12-25 14013.0
Source: Official LLMBoxing leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Llama 2-70B | 22679.0 | Llama 2 70B vs GPT-3.5; archive 2023-12-25 | Jul 2023 | Official leaderboard |
| 2 | Mistral 7B Instruct [snapshot unspecified] | 16615.0 | Llama 2 13B vs Mistral 7B; archive 2023-12-25 | Official leaderboard | |
| 3 | Llama 2-13B | 14029.0 | Llama 2 13B vs Mistral 7B; archive 2023-12-25 | Jul 2023 | Official leaderboard |
| 4 | GPT 3.5 [snapshot unspecified] | 14013.0 | Llama 2 70B vs GPT-3.5; archive 2023-12-25 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official LLMBoxing leaderboard
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Oct 10, 2023 |  Mistral 7B v0.1 technical report | llmboxing.com human evaluation: Mistral 7B preferred 5020 times vs 4143 for Llama 2 13B, as of Oct 6, 2023 (Section 4) | Capabilities |
More in Instructions and chat
- IFEval In use Models answer about 500 prompts containing verifiable instructions such as word limits or required keywords. Code checks each instruction for strict and loose accuracy.
Gemma 4 31B IT 98.9% - IFBench In use 58 new out-of-domain verifiable output constraints. Scored by programmatic constraint checks (strict or loose).
Nemotron-Cascade 2 30B-A3B 82.9% - MultiChallenge In use Multi-turn conversations testing instruction retention, inference memory, versioned editing and self-coherence. Graded by LLM judge with rubrics. GPT-6.1 Sol 82%
- Multi-IF In use Models follow verifiable instructions over three-turn conversations in eight languages, extending IFEval. Scored by instruction accuracy at each turn. LFM3-1.7B 89.1%
- AdvancedIF In use Models respond to over 1,600 prompts with complex, multi-turn, and system-level instructions. A judge model checks each response against expert-written rubrics.
MAI-Thinking-1 85% - IFStruct In use Models generate JSON or YAML that must follow a requested schema and phrasing constraints, without constrained decoding. A response passes only if every structural constraint holds. Gemma 4 31B IT 95.9%