# GPT-6 Astra vs Mistral Large

> GPT-6 Astra is the stronger model overall, scoring 70.3 to 31.6 on the Noometry Index. Mistral Large costs 6.7× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-6-astra-vs-mistral-large
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-6 Astra scores higher in 9 categories and Mistral Large in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Astra leads 93.4 to 17.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for GPT-6 Astra and 8.5% for Mistral Large.
- Mistral Large is cheaper at $2 / $6 per million input/output tokens, against $10 / $50 for GPT-6 Astra.
- GPT-6 Astra accepts more context: 1.05M tokens versus 131K.
- Mistral Large has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Astra | Mistral Large |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 70.3 | 31.6 |
| Rank | 1 | 258 |
| Context | 1.05M | 131K |
| Input $/M | $10 | $2 |
| Output $/M | $50 | $6 |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Astra: 70.5 (#3)
- Mistral Large: 33.8 (#240)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| SciCode | 56.5% | 36.2% |
| LMArena Coding | 1488 | 1277 |
| ALE-Bench | 2,951 | 264.7 |
| DeepSWE | 74.1% | — |
| FrontierCode | 53.3% | — |
| LMArena WebDev | 1792 | — |
| FrontierSWE | 65.5% | — |
| WeirdML | 93.6% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| MirrorCode | 46.7% | — |
| BigCodeBench Complete | — | 38.3% |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |

## Agentic & Tool Use

- GPT-6 Astra: 52.3 (#3)
- Mistral Large: 28.4 (#83)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| APEX-Agents | 64.7% | — |
| Berkeley Function Calling Leaderboard | — | 38.4% |
| Remote Labor Index | 20.8% | — |
| BALROG | 68.3% | — |
| Vending-Bench 2 | 15,515 | — |

## Reasoning

- GPT-6 Astra: 85.1 (#1)
- Mistral Large: 15.4 (#305)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| CritPt | 31.7% | 0% |
| LMArena Hard Prompts | 1464 | 1257 |
| DTBench | 97.3% | 65.1% |
| LMCA | 64.4% | 16.7% |
| Epoch Capabilities Index | 166.6 | 128.52 |
| ARC-AGI-2 | 95% | — |
| SimpleBench | — | 22.5% |
| NYT Connections (extended) | 98.1% | — |
| ARC-AGI-1 | 98.5% | — |
| Chess Puzzles | 72% | — |
| EBR-Bench | 76.2% | — |
| LiveBench Reasoning | — | 43.5% |
| Mystery Game Puzzles | 84% | — |
| LiveBench Data Analysis | — | 50.1% |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |

## Math

- GPT-6 Astra: 93.4 (#1)
- Mistral Large: 17.1 (#285)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 8.5% |
| LMArena Math | 1463 | 1262 |
| FrontierMath (Tiers 1-3) | 93.7% | — |
| FrontierMath Tier 4 | 97.6% | — |
| ProofBench | 99% | — |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
| FrontierMath Erdős | 2.9% | — |

## Knowledge

- GPT-6 Astra: 75.3 (#1)
- Mistral Large: 30.0 (#224)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| GPQA Diamond | 95.8% | 51.3% |
| Vectara Hallucination Rate | 8.7% | 4.5% |
| LMArena Expert | 1486 | 1232 |
| Humanity's Last Exam | 54.8% | — |
| SimpleQA Verified | 75.6% | — |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |

## Multimodal

- GPT-6 Astra: 55.1 (#3)
- Mistral Large: —

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| LMArena Vision | 1280 | — |
| Blueprint-Bench 2 | 49.7% | — |
| Furniture Assembly | 80% | — |
| LMArena Document | 1468 | — |

## Multilingual

- GPT-6 Astra: 54.0 (#53)
- Mistral Large: 40.1 (#214)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| LMArena Non-English | 1431 | 1237 |
| LMArena Chinese | 1495 | 1240 |
| LMArena French | 1467 | 1324 |
| LMArena Russian | 1434 | 1257 |
| LMArena Spanish | 1412 | 1268 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1187 |
| LMArena Korean | — | 1202 |

## Instruction Following

- GPT-6 Astra: 76.6 (#37)
- Mistral Large: 68.1 (#186)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1454 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |

## Long Context

- GPT-6 Astra: 44.7 (#57)
- Mistral Large: 38.3 (#194)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1460 | 1262 |

## Writing & Preference

- GPT-6 Astra: 75.7 (#7)
- Mistral Large: 40.9 (#237)

| Benchmark | GPT-6 Astra | Mistral Large |
|---|---|---|
| LMArena Text | 1443 | 1266 |
| LMArena Creative Writing | 1423 | 1243 |
| EQ-Bench Creative Writing | 2173 | 985 |
| LMArena Multi-Turn | 1457 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |

## FAQ

### Is GPT-6 Astra better than Mistral Large?

GPT-6 Astra is the stronger model overall, scoring 70.3 to 31.6 on the Noometry Index. Mistral Large costs 6.7× less per token, which makes it the better buy when GPT-6 Astra's lead doesn't matter for your workload.

### Which is cheaper, GPT-6 Astra or Mistral Large?

Mistral Large is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-6 Astra lists at $10 and $50.

### Is GPT-6 Astra or Mistral Large better for coding?

GPT-6 Astra scores higher on coding benchmarks: 70.5 versus 33.8 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Astra does, with 1.05M tokens against 131K.

### How many benchmarks do GPT-6 Astra and Mistral Large share?

24 benchmarks have published results for both models. GPT-6 Astra has 51 scored results on Noometry and Mistral Large has 51.