When2Call Multi-Turn (NVIDIA) (When2Call Multi-Turn)
We show this table for reference; we do not rank on it.
NVIDIA multi-turn version of When2Call that tests whether a model calls a tool, asks a follow-up question, or abstains.
Benchmark score on When2Call Multi-Turn — October 29, 2026
We compile the When2Call Multi-Turn rows from provider self-reports. The table contains 2 models. Nemotron 3.5 Bolt E4B BF16 has the highest published score at 63.36%, but the available coverage and evaluation setups do not establish a market leader. We do not use these results to rank models overall.
Nemotron 3.5 Bolt E4B BF16
NVIDIA
Nemotron 3.5 Bolt E4B
NVIDIA
2 modelsAgenticCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (2 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Nemotron 3.5 Bolt E4B BF16NVIDIA | 63.36% | Not reported | Open |
| 2 | Nemotron 3.5 Bolt E4BNVIDIA | 60.50% | Not reported | Open |
How to read this leaderboard
Operator receipt: 2 sourced rows are currently displayable on this page; the highest published score among these 2 models is Nemotron 3.5 Bolt E4B BF16 at 63.36%, which does not establish a market leader.
Honest limit: Unreleased and built by the model's own publisher. Only NVIDIA-run results are stored so far, from one harness. Display-only and excluded from weighted ranking.
About When2Call Multi-Turn
Year
2026
Tasks
Multi-turn tool-use decisions (not yet published)
Format
Overall accuracy
Difficulty
When to call, clarify or abstain
NVIDIA reports it for Nemotron 3.5 Bolt E4B (October 23, 2026), run in its NeMo Gym harness, as overall accuracy in reasoning mode. NVIDIA says this multi-turn version will be released later, so its tasks and scoring are not public yet.
Freshness and provenance
Version
When2Call Multi-Turn 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does When2Call Multi-Turn measure?
NVIDIA multi-turn version of When2Call that tests whether a model calls a tool, asks a follow-up question, or abstains.
Which model scores highest on When2Call Multi-Turn?
Nemotron 3.5 Bolt E4B BF16 by NVIDIA currently leads with a score of 63.36% on When2Call Multi-Turn.
How many models are evaluated on When2Call Multi-Turn?
2 AI models have published results on When2Call Multi-Turn in the BenchLM catalog.
Compare top models on When2Call Multi-Turn
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.