Across 27 comparable benchmarks, GPT-6 Astra has better reported results on 23 and worse on 4. 2 further shared results are listed below, each with the reason it is not counted.
Moderate evidence27 of 29 shared results countedHow we compare
Price basis: Standard rates vs listed rates. The price section below shows the applicable periods.
| Fact | Mistral Large 4 | GPT-6 Astra |
|---|---|---|
| Context window | 1M tokens | 1.05M tokens |
| Input / 1M | $0.700 | $10.00 |
| Output / 1M | $2.10 | $50.00 |
| Blended (4:1) | $0.526 | $11.52 |
| Status | Preview | Preview |
| Released | October 6, 2026 | September 3, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates β set your own mix.
Mean of 18 shared benchmarks grouped by capability on a 0β100 scale. How averages are built
GPT-6 Astra ahead on 8 of 10 counted
GPT-6 Astra ahead on 3 of 3 counted
GPT-6 Astra ahead on 4 of 4 counted
GPT-6 Astra ahead on 1 of 1 counted
GPT-6 Astra ahead on 3 of 5 counted
No shared comparable results for Abstract reasoning, Computer & tool use, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show Mistral Large 4 relative to GPT-6 Astra. Reading the results
| Benchmark | Mistral Large 4 | GPT-6 Astra | Gap |
|---|---|---|---|
| AA Cyber IndexRepository Engineering Β· Composite index (0β100) | 49.5Thinking Β· tools | 33.4β Max Β· tools | 1.48Γ#7 vs #15 of 22 |
| Code MigrationMaintenance & Optimization Β· Hidden-test pass rate (%) | 30.6Thinking High Β· tools | 67.7β Max Β· tools | β37.2 pts#43 vs #4 of 79 |
| DeepSWERepository Engineering Β· Pass@1 (DeepSWE v1.1) | 61.7Thinking Β· tools | 74.1β Extra-High Β· tools | β12.4 pts#45 vs #13 of 73 |
| FrontierCode 1.1Code Generation & Editing Β· Weighted rubric score (%)Not counted Β· Different or unrecorded test environments β excluded from the tally | 39.4Thinking High Β· toolsΒ· frontiercode-1-1-main | 64.5β Max Β· toolsΒ· frontiercode-1-1-extended | β#30 vs #2 of 51 |
| FrontierCode 2.0Repository Engineering Β· Weighted rubric score | 8.8Thinking High Β· tools + web | 44.0β Max Β· tools + web | 0.20Γ#30 vs #6 of 36 |
| IOI (Vals v2)Algorithmic Coding Β· Subtask points (%) | 45.3Thinking High Β· tools | 100β Max Β· tools | β54.7 pts#32 vs #1 of 45 |
| SWE-Atlas-QnARepository Engineering Β· Task Resolve Rate (%) | 59.4Thinking Β· tools | 59.1β Extra-High Β· tools | +0.3 pts#4 vs #6 of 17 |
| Terminal-Bench 4.0Agentic Development Β· Resolution rate (%)Not counted Β· Different or unrecorded test environments β excluded from the tally | 28.3Thinking Β· toolsΒ· legacy-default | 59.6β Unspecified Reasoning Β· toolsΒ· vals-mini-swe | β#40 vs #11 of 90 |
| Terminal-Bench 4.1Agentic Development Β· Resolution rate (%) | 27.6Max Β· tools | 60.3Max Β· tools | β32.7 pts#26 vs #8 of 48 |
| Vibe Code Bench 1-100Agentic Development Β· Accuracy (%) | 14.3Thinking High Β· tools | 27.6β Max Β· tools | β13.4 pts#20 vs #5 of 25 |
| Vibe Code Bench v1.1Code Generation & Editing Β· Pass rate (%) | 78.4Thinking High Β· tools | 89.6β Max Β· tools | β11.2 pts#36 vs #10 of 121 |
| WeirdML v3ML Engineering Β· Official score (0-1; higher is better) | 0.044Thinking High Β· tools | 0.42β Extra-High Β· tools | 0.11Γ#25 vs #1 of 26 |
| FrontierMath Tier 4 v2Mathematics Β· Accuracy (verification_code) | 26.8Thinking High Β· tools | 97.6Thinking High Β· tools | β70.8 pts#40 vs #4 of 73 |
| FrontierMath v2Mathematics Β· Accuracy (verification_code) | 58.6Thinking High Β· tools | 93.7β Max Β· tools | β35.1 pts#46 vs #2 of 85 |
| ProofBench v1.1 (Lean 4)Mathematics Β· Accuracy (%) | 10.0Thinking High Β· tools | 99.0β Max Β· tools | β89.0 pts#49 vs #6 of 54 |
| MysteryMechanismScientific Reasoning Β· Accuracy (%) | 12.2Thinking High Β· tools | 53.1β Max Β· tools | β41.0 pts#32 vs #2 of 32 |
| BioMysteryBenchBiology & Genomics Β· Accuracy (%) | 67.0Thinking High Β· tools | 79.3β Max Β· tools | β12.2 pts#21 vs #8 of 28 |
| MedCodeClinical Workflows Β· Accuracy (%) | 40.7Thinking High Β· no tools | 48.5β Max Β· tools | β7.8 pts#58 vs #29 of 97 |
| MedScribeClinical Workflows Β· Rubric pass rate (%) | 80.4Thinking High Β· no tools | 87.9β Max Β· tools | β7.5 pts#53 vs #16 of 99 |
| MMMU-ProVisual Understanding Β· Accuracy | 76.4Thinking Β· no tools | 87.8β Max Β· tools | β11.4 pts#55 vs #5 of 136 |
| EMB (Excel Modeling)Finance Β· Accuracy (%) | 56.2Thinking High Β· tools | 71.7β Max Β· tools | β15.5 pts#49 vs #13 of 78 |
| Finance Agent v2Finance Β· Score | 54.7Thinking High Β· tools | 53.5β Max Β· tools | 1.02Γ#24 vs #32 of 75 |
| Harvey's Legal Agent BenchmarkLegal Β· Accuracy (%) | 15.8Thinking High Β· tools | 5.4β Max Β· tools | +10.4 pts#6 vs #26 of 52 |
| Legal Research BenchLegal Β· Score | 31.7Thinking High Β· tools | 39.4β Max Β· tools | 0.80Γ#52 vs #38 of 91 |
| Tax Agent BenchFinance Β· Accuracy (%, rubric Γ citation quality) | 63.3Thinking High Β· tools | 63.3β Max Β· tools | β0.1 pts#14 vs #13 of 74 |
| AA Intelligence Index v5.0.1Capability Indices Β· Index (0-100) | 25.3Thinking Β· tools | 56.9β Max Β· task tools | 0.45Γ#39 vs #7 of 62 |
| LiveBenchCross-capability Suites Β· Accuracy | 71.8Thinking High Β· no tools | 82.2β Max Β· no tools | β10.3 pts#53 vs #4 of 106 |
| LiveBench 2.0Cross-capability Suites Β· Overall score (equal-weight mean of 7 suite scores, 0-100) | 47.7Thinking High Β· tools | 86.1β Max Β· task tools | 0.55Γ#47 vs #3 of 51 |
| Vals Index v2.1Capability Indices Β· GDP-weighted accuracy (%) | 48.0Thinking High Β· tools | 63.1β Max Β· task tools | β15.1 pts#30 vs #6 of 38 |
Repository Engineering Β· Composite index (0β100)
Maintenance & Optimization Β· Hidden-test pass rate (%)
Repository Engineering Β· Pass@1 (DeepSWE v1.1)
Code Generation & Editing Β· Weighted rubric score (%)
Not counted Β· Different or unrecorded test environments β excluded from the tally
Repository Engineering Β· Weighted rubric score
Algorithmic Coding Β· Subtask points (%)
Repository Engineering Β· Task Resolve Rate (%)
Agentic Development Β· Resolution rate (%)
Not counted Β· Different or unrecorded test environments β excluded from the tally
Agentic Development Β· Resolution rate (%)
Agentic Development Β· Accuracy (%)
Code Generation & Editing Β· Pass rate (%)
ML Engineering Β· Official score (0-1; higher is better)
Mathematics Β· Accuracy (verification_code)
Mathematics Β· Accuracy (verification_code)
Mathematics Β· Accuracy (%)
Scientific Reasoning Β· Accuracy (%)
Biology & Genomics Β· Accuracy (%)
Clinical Workflows Β· Accuracy (%)
Clinical Workflows Β· Rubric pass rate (%)
Visual Understanding Β· Accuracy
Finance Β· Accuracy (%)
Finance Β· Score
Legal Β· Accuracy (%)
Legal Β· Score
Finance Β· Accuracy (%, rubric Γ citation quality)
Capability Indices Β· Index (0-100)
Cross-capability Suites Β· Accuracy
Cross-capability Suites Β· Overall score (equal-weight mean of 7 suite scores, 0-100)
Capability Indices Β· GDP-weighted accuracy (%)
A dash means no result is available.
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix Β· 5% steps
Blended price = effective input rate at 90% cache hits Γ 80% + output rate Γ 20%
Mistral Large 4 costs95%lessthan GPT-6 Astra per million tokens β 21.9Γ cheaper.Using standard rates vs listed rates at this token mix.
Mistral Large 4: Standard pricing. Mistral official pricing (docs.mistral.ai/inference/pricing) verified 2026-11-13; price reduced from $1.36 Mistral official pricing (docs.mistral.ai/inference/pricing) verified 2026-11-13; price reduced from $4.18
GPT-6 Astra: Standard pricing. context_length <= 272000 OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-11-13; standard tier; <=272K input tokens context_length > 272000 OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-11-13; standard tier; >272K input tokens
| Price / 1M tokens | Mistral Large 4 | GPT-6 Astra | Ratio |
|---|---|---|---|
| Input | $0.70 | $10.00 | 0.07Γ |
| Output | $2.10 | $50.00 | 0.04Γ |
| Cached input | $0.07 | $1.00 | 0.07Γ |
| Cache write | β | $12.50 | β |
| Blended | $0.526 | $11.52 | 0.05Γ |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
Mistral Large 4 relative to GPT-6 Astra
Summary: 4 of 8 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 4 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Mistral Large 4 4.5Γ cheaper
Mistral Large 4 3.5Γ cheaper
Mistral Large 4 2.8Γ cheaper
Mistral Large 4 1.9Γ cheaper
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Mistral Large 4 4.5Γ cheaper
Mistral Large 4 3.5Γ cheaper
Mistral Large 4 2.8Γ cheaper
Mistral Large 4 1.9Γ cheaper
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Large 4
GPT-6 Astra
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Large 4
GPT-6 Astra
Settings or sources differ or are not recorded; interpret the comparison with care.
Mistral Large 4
GPT-6 Astra
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Large 4
GPT-6 Astra
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Large 4
GPT-6 Astra
Settings or sources differ or are not recorded; interpret the comparison with care.
Mistral Large 4
GPT-6 Astra
Settings or sources differ or are not recorded; interpret the comparison with care.
Mistral Large 4
GPT-6 Astra
Settings or sources differ or are not recorded; interpret the comparison with care.
Mistral Large 4
GPT-6 Astra
| Benchmark | Mistral Large 4 | GPT-6 Astra |
|---|---|---|
| AA Coding Agent Index v2.0 | Not reported | |
| AA Intelligence Index v4.3 | Not reported | 2 more settings |
| AA-Omniscience | Not reported | 2 more settings |
| Agent Arena | Not reported | $5.4265 / task Median Β· Net improvement index (percentage points vs. average model): 14.6 Max (With Tools + Internet) |
| Agent Arena Index | Not reported | $4.7264 / task Median Β· Agent Arena Index rating (index points): 1282 Max (With Tools + Internet) |
| Agents' Last Exam v1.0 | Not reported | 3 more settings |
| ARC-AGI-1 | Not reported | 5 more settings$0.2837 / task Average Β· Score (%): 98.5 Thinking High (With Tools) $0.2265 / task Average Β· Score (%): 96.5 Thinking Low (With Tools) $0.2398 / task Average Β· Score (%): 97.5 Thinking Medium (With Tools) |
| ARC-AGI-2 | Not reported | $0.6681 / task Average Β· ARC-AGI-2 Score: 92.1 Thinking High (With Tools) 3 more settings$0.4162 / task Average Β· ARC-AGI-2 Score: 85.4 Thinking Low (With Tools) $0.4801 / task Average Β· ARC-AGI-2 Score: 92.1 Thinking Medium (With Tools) $0.8287 / task Average Β· ARC-AGI-2 Score: 93.3 Extra-High (With Tools) |
| AutomationBench | Not reported | $1.44 / task Average Β· Pass Rate: 37.1 Thinking High (With Tools) Β· zapier-private-heldout 3 more settings$1.08 / task Average Β· Pass Rate: 30.3 Thinking Low (With Tools) Β· zapier-private-heldout $1.27 / task Average Β· Pass Rate: 34.1 Thinking Medium (With Tools) Β· zapier-private-heldout $1.50 / task Average Β· Pass Rate: 39.0 Extra-High (With Tools) Β· zapier-private-heldout |
| BenchCAD 2.1 Core | Not reported | $9.71 / task Average Β· Core mean geometry/circuit score (0β100): 62.6 Max (With Tools) $5.38 / task Average Β· Core mean geometry/circuit score (0β100): 51.1 Thinking High (With Tools) $1.14 / task Average Β· Core mean geometry/circuit score (0β100): 20.3 Thinking Low (With Tools) 2 more settings$3.73 / task Average Β· Core mean geometry/circuit score (0β100): 38.7 Thinking Medium (With Tools) $6.32 / task Average Β· Core mean geometry/circuit score (0β100): 53.9 Extra-High (With Tools) |
| BioMysteryBench | Not reported | |
| Code Migration | Not reported | |
| CWE-bench v1 | Not reported | |
| DeepSWE | Not reported | 2 more settings |
| EMB (Excel Modeling) | Not reported | |
| Finance Agent v2 | Not reported | |
| FrontierCode 1.1 | Not reported | $4.586 / task Average Β· Weighted rubric score: 53.3 Max (With Tools) Β· frontiercode-1-1-main $3.0149 / task Average Β· Weighted rubric score: 50.9 Thinking High (With Tools) Β· frontiercode-1-1-main $1.70 / task Average Β· Weighted rubric score: 45.3 Thinking Low (With Tools) Β· frontiercode-1-1-main 2 more settings$2.4259 / task Average Β· Weighted rubric score: 48.8 Thinking Medium (With Tools) Β· frontiercode-1-1-main $3.2755 / task Average Β· Weighted rubric score: 50.6 Extra-High (With Tools) Β· frontiercode-1-1-main |
| FrontierCS 2 | Not reported | |
| FrontierSWE v2 | Not reported | |
| GDPx | Not reported | |
| Harvey's Legal Agent Benchmark | Not reported | |
| HLE | Not reported | $0.0389 / task Average Β· Accuracy: 49.2 Thinking Low (No Tools) Β· Text only Β· text-only $0.1005 / task Average Β· Accuracy: 52.7 Thinking Medium (No Tools) Β· Text only Β· text-only 2 more settings$0.1618 / task Average Β· Accuracy: 53.1 Thinking High (No Tools) Β· Text only Β· text-only $0.2532 / task Average Β· Accuracy: 54.6 Extra-High (No Tools) Β· Text only Β· text-only |
| IOI (Vals v2) | Not reported | |
| Legal Research Bench | Not reported | |
| MedCode | Not reported | |
| MedScribe | Not reported | |
| MysteryMechanism | Not reported | |
| ProofBench v1.1 (Lean 4) | Not reported | |
| SWE-sweep | Not reported | |
| Tax Agent Bench | $2.7212 / task Average Β· Accuracy (%, rubric Γ citation quality): 63.3 Thinking High (With Tools) | Not reported |
| Terminal-Bench-Science 0.1 | Not reported | $20.80 / task Average Β· Task success rate (%): 62.9 Unspecified Reasoning (With Tools) |
| Vals Index v2.2 | Not reported | |
| Vibe Code Bench v1.1 | Not reported |
| Attribute | Mistral Large 4 | GPT-6 Astra |
|---|---|---|
| Availability | Preview | Preview |
| Provider | MistralAI | OpenAI |
| Model type | Multimodal model | Reasoning model |
| First release | October 6, 2026 | September 3, 2026 |
| Context window | 1M tokens | 1.05M tokens |
| Max output | β | 128K tokens |
| Total parameters | 1.1T | β |
| Active parameters | 49B | β |
| Architecture | Mixture of Experts (MoE) | Undisclosed |
| Documented capabilities | Reasoning, Multilingual, Function calling, Tool use, Structured output, Long context, Coding | Reasoning, Multilingual, Coding, Function calling, Tool use, Structured output, Computer use, Web search, Long context |
| Knowledge cutoff | β | 2026-04-30 |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Weights | Closed / API only | Closed / API only |
| Commercial use | Proprietary | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used Mistral Large 4 or GPT-6 Astra? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.