# EQ-Bench 3 leaderboard

Instruction Following benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/leaderboards/benchmarks/instruction-following-eq-bench3

Every model the catalog carries a published EQ-Bench 3 value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Instruction Following |
| Measure | LLM-judged rubric score |
| Tasks | Multi-turn roleplay scenarios |
| Difficulty | Emotional and social intelligence |
| Published by | [EQ-Bench 3](https://eqbench.com/eqbench3.html) |

Measures emotional intelligence in multi-turn roleplay scenarios, scored by an LLM judge against a rubric.

## EQ-Bench 3 ranking

2 models with a published EQ-Bench 3 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | LLM-judged rubric score | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Nemotron 3.5 Bolt E4B BF16](https://modelscale.dev/models/nemotron-3-5-bolt-e4b-bf16) | NVIDIA | 39.40 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Nemotron 3.5 Bolt E4B](https://modelscale.dev/models/nemotron-3-5-bolt-e4b) | NVIDIA | 38.15 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
