Across 27 comparable benchmarks, Claude Opus 5 has better reported results on 26 and worse on 1. 1 further shared result is listed below, each with the reason it is not counted.
Moderate evidence27 of 28 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Claude Opus 5 | Mistral Large 4 |
|---|---|---|
| Context window | 1M tokens | 1M tokens |
| Input / 1M | $5.00 | $0.700 |
| Output / 1M | $25.00 | $2.10 |
| Blended (4:1) | $5.76 | $0.526 |
| Status | Available | Preview |
| Released | July 24, 2026 | October 6, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates β set your own mix.
Mean of 19 shared benchmarks grouped by capability on a 0β100 scale. How averages are built
Claude Opus 5 ahead on 10 of 10 counted
Claude Opus 5 ahead on 3 of 3 counted
Claude Opus 5 ahead on 4 of 4 counted
Claude Opus 5 ahead on 1 of 1 counted
Claude Opus 5 ahead on 4 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 Claude Opus 5 relative to Mistral Large 4. Reading the results
| Benchmark | Claude Opus 5 | Mistral Large 4 | Gap |
|---|---|---|---|
| Code MigrationMaintenance & Optimization Β· Hidden-test pass rate (%) | 57.5Max Β· tools | 30.6β Thinking High Β· tools | +26.9 pts#9 vs #43 of 79 |
| DeepSWERepository Engineering Β· Pass@1 (DeepSWE v1.1) | 73.7Max Β· tools | 61.7β Thinking Β· tools | +12.0 pts#16 vs #45 of 73 |
| FrontierCode 1.1Code Generation & Editing Β· Weighted rubric score (%)Not counted Β· Different or unrecorded test environments β excluded from the tally | 63.6Unspecified Reasoning Β· toolsΒ· frontiercode-1-1-extended | 39.4β Thinking High Β· toolsΒ· frontiercode-1-1-main | β#3 vs #30 of 51 |
| FrontierCode 2.0Repository Engineering Β· Weighted rubric score | 26.7Extra-High Β· tools + internet | 8.8β Thinking High Β· tools + web | 3.03Γ#15 vs #30 of 36 |
| IOI (Vals v2)Algorithmic Coding Β· Subtask points (%) | 84.3Max Β· tools | 45.3β Thinking High Β· tools | +39.0 pts#8 vs #32 of 45 |
| SWE-Atlas-QnARepository Engineering Β· Task Resolve Rate (%) | 63.2Extra-High Β· tools | 59.4β Thinking Β· tools | +3.8 pts#2 vs #4 of 17 |
| Terminal-Bench 4.0Agentic Development Β· Resolution rate (%) | 53.9Extra-High Β· tools | 28.3β Thinking Β· tools | +25.6 pts#16 vs #40 of 90 |
| Terminal-Bench 4.1Agentic Development Β· Resolution rate (%) | 53.3Extra-High Β· tools | 27.6β Max Β· tools | +25.8 pts#13 vs #26 of 48 |
| Vibe Code Bench 1-100Agentic Development Β· Accuracy (%) | 28.5Max Β· tools | 14.3β Thinking High Β· tools | +14.3 pts#3 vs #20 of 25 |
| Vibe Code Bench v1.1Code Generation & Editing Β· Pass rate (%) | 88.4Max Β· tools | 78.4β Thinking High Β· tools | +10.0 pts#13 vs #36 of 121 |
| WeirdML v3ML Engineering Β· Official score (0-1; higher is better) | 0.19Extra-High Β· tools | 0.044β Thinking High Β· tools | 4.28Γ#9 vs #25 of 26 |
| FrontierMath Tier 4 v2Mathematics Β· Accuracy (verification_code) | 73.2Max Β· tools | 26.8β Thinking High Β· tools | +46.3 pts#15 vs #40 of 73 |
| FrontierMath v2Mathematics Β· Accuracy (verification_code) | 85.6Max Β· tools | 58.6β Thinking High Β· tools | +27.0 pts#15 vs #46 of 85 |
| ProofBench v1.1 (Lean 4)Mathematics Β· Accuracy (%) | 99.0Max Β· tools | 10.0β Thinking High Β· tools | +89.0 pts#6 vs #49 of 54 |
| MysteryMechanismScientific Reasoning Β· Accuracy (%) | 37.4Max Β· tools | 12.2β Thinking High Β· tools | +25.2 pts#10 vs #32 of 32 |
| BioMysteryBenchBiology & Genomics Β· Accuracy (%) | 79.3Max Β· tools | 67.0β Thinking High Β· tools | +12.2 pts#8 vs #21 of 28 |
| MedCodeClinical Workflows Β· Accuracy (%) | 63.6Max Β· tools | 40.7β Thinking High Β· no tools | +22.9 pts#1 vs #58 of 97 |
| MedScribeClinical Workflows Β· Rubric pass rate (%) | 91.0Max Β· tools | 80.4β Thinking High Β· no tools | +10.5 pts#4 vs #53 of 99 |
| MMMU-ProVisual Understanding Β· Accuracy | 84.7Max Β· no tools | 76.4β Thinking Β· no tools | +8.3 pts#10 vs #55 of 136 |
| EMB (Excel Modeling)Finance Β· Accuracy (%) | 73.6Max Β· tools | 56.2β Thinking High Β· tools | +17.4 pts#8 vs #49 of 78 |
| Finance Agent v2Finance Β· Score | 58.6Max Β· tools | 54.7β Thinking High Β· tools | 1.07Γ#9 vs #24 of 75 |
| Harvey's Legal Agent BenchmarkLegal Β· Accuracy (%) | 6.7Max Β· tools | 15.8β Thinking High Β· tools | β9.2 pts#21 vs #6 of 52 |
| Legal Research BenchLegal Β· Score | 55.3Max Β· tools | 31.7β Thinking High Β· tools | 1.74Γ#3 vs #52 of 91 |
| Tax Agent BenchFinance Β· Accuracy (%, rubric Γ citation quality) | 75.1Max Β· tools | 63.3β Thinking High Β· tools | +11.8 pts#2 vs #14 of 74 |
| AA Intelligence Index v5.0.1Capability Indices Β· Index (0-100) | 44.2Max Β· task tools | 25.3β Thinking Β· tools | 1.74Γ#15 vs #39 of 62 |
| LiveBenchCross-capability Suites Β· Accuracy | 80.1Max Β· no tools | 71.8β Thinking High Β· no tools | +8.2 pts#11 vs #53 of 106 |
| LiveBench 2.0Cross-capability Suites Β· Overall score (equal-weight mean of 7 suite scores, 0-100) | 82.1Max Β· task tools | 47.7β Thinking High Β· tools | 1.72Γ#8 vs #47 of 51 |
| Vals Index v2.1Capability Indices Β· GDP-weighted accuracy (%) | 63.7Max Β· task tools | 48.0β Thinking High Β· tools | +15.6 pts#5 vs #30 of 38 |
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 (%)
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%
Claude Opus 5 costs994%morethan Mistral Large 4 per million tokens β 10.9Γ the price.Using standard rates at this token mix.
Claude Opus 5: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-10-21; full 1M context at standard price
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
| Price / 1M tokens | Claude Opus 5 | Mistral Large 4 | Ratio |
|---|---|---|---|
| Input | $5.00 | $0.70 | 7.14Γ |
| Output | $25.00 | $2.10 | 11.9Γ |
| Cached input | $0.50 | $0.07 | 7.14Γ |
| Cache write | β | β | β |
| Blended | $5.76 | $0.526 | 10.9Γ |
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.
Claude Opus 5 relative to Mistral Large 4
Summary: 3 of 9 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 3 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Claude Opus 5 8.3Γ more expensive
Claude Opus 5 6.8Γ more expensive
Claude Opus 5 5.5Γ more expensive
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Claude Opus 5 8.3Γ more expensive
Claude Opus 5 6.8Γ more expensive
Claude Opus 5 5.5Γ more expensive
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5
Mistral Large 4
Effort, protocol, aggregation and source match; other run conditions may differ.
Claude Opus 5
Mistral Large 4
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5
Mistral Large 4
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5
Mistral Large 4
Effort, protocol, aggregation and source match; other run conditions may differ.
Claude Opus 5
Mistral Large 4
Effort, protocol, aggregation and source match; other run conditions may differ.
Claude Opus 5
Mistral Large 4
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5
Mistral Large 4
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5
Mistral Large 4
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5
Mistral Large 4
| Benchmark | Claude Opus 5 | Mistral Large 4 |
|---|---|---|
| AA Intelligence Index v4.3 | 2 more settings | Not reported |
| AA-Omniscience | 2 more settings | Not reported |
| Agent Arena | $4.6032 / task Median Β· Net improvement index (percentage points vs. average model): 7.1 Max (With Tools + Internet) | Not reported |
| Agent Arena Index | $4.6109 / task Median Β· Agent Arena Index rating (index points): 1176 Max (With Tools + Internet) | Not reported |
| Agents' Last Exam v1.0 | 2 more settings | Not reported |
| ARC-AGI-1 | Not reported | |
| ARC-AGI-2 | $1.45 / task Average Β· ARC-AGI-2 Score: 88.3 Thinking High (With Tools) | Not reported |
| AutomationBench | $1.03 / task Average Β· Pass Rate: 20.6 Thinking High (With Tools) Β· zapier-private-heldout $0.75 / task Average Β· Pass Rate: 20.4 Thinking Low (With Tools) Β· zapier-private-heldout 2 more settings$0.89 / task Average Β· Pass Rate: 23.9 Thinking Medium (With Tools) Β· zapier-private-heldout $1.15 / task Average Β· Pass Rate: 25.3 Extra-High (With Tools) Β· zapier-private-heldout | Not reported |
| BioMysteryBench | Not reported | |
| Code Migration | Not reported | |
| CursorBench 4.0 | $11.95 / task Average Β· Task score: 46.6 Max (With Tools) Β· cursorbench-4-0-cursor-agent $9.00 / task Average Β· Task score: 44.7 Thinking High (With Tools) Β· cursorbench-4-0-cursor-agent $4.87 / task Average Β· Task score: 40.7 Thinking Low (With Tools) Β· cursorbench-4-0-cursor-agent 2 more settings$6.94 / task Average Β· Task score: 43.3 Thinking Medium (With Tools) Β· cursorbench-4-0-cursor-agent $11.43 / task Average Β· Task score: 46.1 Extra-High (With Tools) Β· cursorbench-4-0-cursor-agent | Not reported |
| DeepSWE | 2 more settings | Not reported |
| EMB (Excel Modeling) | Not reported | |
| Finance Agent v2 | Not reported | |
| FrontierCode 1.1 | $11.416 / task Average Β· Weighted rubric score: 48.0 Max (With Tools) Β· frontiercode-1-1-main $7.2397 / task Average Β· Weighted rubric score: 48.0 Thinking High (With Tools) Β· frontiercode-1-1-main $2.678 / task Average Β· Weighted rubric score: 42.0 Thinking Low (With Tools) Β· frontiercode-1-1-main 2 more settings$4.3127 / task Average Β· Weighted rubric score: 53.4 Thinking Medium (With Tools) Β· frontiercode-1-1-main $9.1423 / task Average Β· Weighted rubric score: 43.6 Extra-High (With Tools) Β· frontiercode-1-1-main | Not reported |
| FrontierSWE v2 | Not reported | |
| Harvey's Legal Agent Benchmark | Not reported | |
| HLE | $0.0625 / task Average Β· Accuracy: 43.4 Thinking Low (No Tools) Β· Text only Β· text-only $0.187 / task Average Β· Accuracy: 51.3 Thinking Medium (No Tools) Β· Text only Β· text-only 2 more settings$0.3521 / task Average Β· Accuracy: 52.8 Thinking High (No Tools) Β· Text only Β· text-only $0.5171 / task Average Β· Accuracy: 54.4 Extra-High (No Tools) Β· Text only Β· text-only | Not reported |
| 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 | |
| Vibe Code Bench v1.1 | Not reported |
| Attribute | Claude Opus 5 | Mistral Large 4 |
|---|---|---|
| Availability | Available | Preview |
| Provider | Anthropic | MistralAI |
| Model type | Reasoning model | Multimodal model |
| First release | July 24, 2026 | October 6, 2026 |
| Context window | 1M tokens | 1M tokens |
| Max output | 128K tokens | β |
| Total parameters | β | 1.1T |
| Active parameters | β | 49B |
| Architecture | Undisclosed | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual, Coding, Tool use, Function calling, Computer use, Long context | Reasoning, Multilingual, Function calling, Tool use, Structured output, Long context, Coding |
| Knowledge cutoff | 2026-05 | β |
| 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 Claude Opus 5 or Mistral Large 4? 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.