MosaicMLMPT-30B
MPT-30B is a MosaicML proprietary model, released 22 Jun 2023. It is not yet ranked: 4 independent results so far, and the index needs at least five across two categories. Last measured 22 Oct 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- –
- Full answer
- –
- Cost per run
- –
- Released
- 22 Jun 2023
- Reasoning28
- Coding26
- Maths24
- Human preference27
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1158#334 | Arena (LMArena) | |
| BIG-Bench Hardnot in index | 38.0%#32 | Epoch AI Benchmarking Hub | |
| WinoGrandenot in index | 71.0%#43 | Epoch AI Benchmarking Hub | |
| ARC (AI2 Challenge)not in index | 50.6%#47 | Epoch AI Benchmarking Hub | |
| PIQAnot in index | 81.9%#23 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1168#338 | Arena (LMArena) |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Maths | 1096#338 | Arena (LMArena) | |
| GSM8Knot in index | 34.4%#49 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| MMLUnot in index | 47.9%#94 | Epoch AI Benchmarking Hub | |
| TriviaQAnot in index | 73.6%#25 | Epoch AI Benchmarking Hub | |
| OpenBookQAnot in index | 52.0%#24 | Epoch AI Benchmarking Hub |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Instruction Followingnot in index | 1135#330 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1151#336 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1132#333 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1126#329 | Arena (LMArena) |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 100.3#220 | Epoch AI Benchmarking Hub |
More from MosaicML
See also
Questions people ask
- Why does MPT-30B have no BenchLeader Index score?
- The index needs at least five independent results across at least two categories; as of 26 Oct 2026 MPT-30B has 4 that count. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
- When was MPT-30B released?
- MosaicML released MPT-30B on 22 Jun 2023.
Data as of 26 Oct 2026. Compare with another model.
Cite as: BenchLeader, “MPT-30B: benchmarks, pricing, speed and rank”, data as of 26 Oct 2026, https://www.benchleader.com/models/mpt-30b.