IBM
availableShows if the model has enough results for an index.Granite 4.2 30B
Granite 4.2 30B is a reasoning model from IBM in the Granite 4.2 family. 20 benchmarks count toward its score, in 6 categories.
IndexOverall score out of 100.41.9 ±5.1
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.—
Input / 1MUS dollars per 1M input tokens.Free
Output / 1MUS dollars per 1M output tokens.Free
ContextMaximum tokens in one request.128K
EloLMArena rating and rank.1363 (#181)
The index is a score out of 100. The ± range shows how much it can change.
3,255 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category, out of 100.
Out of 100Results
20 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result as a score out of 100. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| American Invitational Mathematics Examination 2025 | Math | 89.2% | 47.0 | — | — | Mathematical Association of America |
| Harvard-MIT Mathematics Tournament February 2025 | Math | 89.2% | 48.3 | — | — | Qwen |
| Massive Multitask Language Understanding Professional | Knowledge | 77.6% | 42.8 | — | — | Yubo Wang et al. |
| Instruction Following Benchmark | Instruction | 77.2% | 55.8 | — | — | Benchmark authors |
| LiveCodeBench v6 | Coding | 75.8% | 46.0 | — | — | LiveCodeBench maintainers |
| Graduate-Level Google-Proof Q&A | Knowledge | 66.4% | 39.6 | — | — | David Rein et al. |
| Artificial Analysis GPQA Diamond | Knowledge | 64.4% | 34.6 | — | — | Artificial Analysis |
| τ³-Bench Tool-Agent-User Evaluation | Agentic | 62.0% | 47.9 | — | — | Sierra Research |
| Berkeley Function Calling Leaderboard v4 | Agentic | 61.4% | 46.1 | — | — | Arcee AI |
| Software Engineering Benchmark Verified | Coding | 57.0% | 43.2 | — | — | Carlos E. Jimenez et al. |
| Artificial Analysis Long Context Reasoning | Reasoning | 49.0% | 42.2 | — | — | Artificial Analysis |
| Scientific Code Benchmark | Coding | 38.8% | 48.6 | — | — | Benchmark authors |
| Artificial Analysis SciCode | Coding | 37.8% | 45.3 | — | — | Artificial Analysis |
| SWE-bench Pro | Coding | 33.3% | 36.2 | — | — | Xiang Deng et al. |
| Terminal-Bench 2.1 (provider run) | Agentic | 29.2% | 41.3 | — | — | DeepSeek-AI |
| Terminal-Bench 2.1 (provider run) | Agentic | 29.2% | 41.3 | — | — | DeepSeek-AI |
| Artificial Analysis Intelligence Index | Knowledge | 14.8% | 40.9 | — | — | Artificial Analysis |
| Artificial Analysis Humanity's Last Exam | Knowledge | 11.2% | 36.7 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 10.1% | 26.4 | — | — | Artificial Analysis |
| GDPval-AA normalized | Agentic | 3.2% | 38.0 | — | — | Artificial Analysis |
| Critical Physics Tasks | Reasoning | 0.3% | 39.0 | — | — | Artificial Analysis |
20 benchmarks count, from 21 of 21 results. A grey row does not count. Too few models took that benchmark.