Across 18 comparable benchmarks, Claude Opus 5.5 has better reported results on 18 and worse on 0. 1 further shared result is listed below, each with the reason it is not counted.
Moderate evidence18 of 19 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Claude Opus 5.5 | Mistral Medium 3.5 |
|---|---|---|
| Context window | 1M tokens | 256K tokens |
| Input / 1M | $4.00 | $1.50 |
| Output / 1M | $20.00 | $7.50 |
| Blended (4:1) | $4.46 | $2.70 |
| Status | Available | Available |
| Released | September 22, 2026 | May 1, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 13 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Claude Opus 5.5 ahead on 3 of 3 counted
Claude Opus 5.5 ahead on 1 of 1 counted
Claude Opus 5.5 ahead on 3 of 3 counted
Claude Opus 5.5 ahead on 2 of 2 counted
Claude Opus 5.5 ahead on 8 of 8 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 5.5 relative to Mistral Medium 3.5. Reading the results
| Benchmark | Claude Opus 5.5 | Mistral Medium 3.5 | Gap |
|---|---|---|---|
| Code MigrationMaintenance & Optimization · Hidden-test pass rate (%) | 66.7Max · tools | 5.1≠Thinking High · tools | +61.5 pts#5 vs #72 of 79 |
| FrontierCode 1.1Code Generation & Editing · Weighted rubric score (%) | 54.6Thinking Medium · tools· frontiercode-1-1-main | 8.0≠Normal · tools· frontiercode-1-1-main | +46.6 pts#1 vs #51 of 51 |
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 90.3Max · tools | 2.9≠Thinking High · tools | +87.4 pts#7 vs #115 of 121 |
| ProofBench v1.1 (Lean 4)Mathematics · Accuracy (%) | 100.0Max · tools | 9.0≠Thinking High · tools | +91.0 pts#1 vs #50 of 54 |
| SciCodeScientific Computing · Score | 66.9Max · no tools | 40.2≠Thinking · no tools | 1.67×#2 vs #75 of 98 |
| MedCodeClinical Workflows · Accuracy (%) | 49.8Max · tools | 33.8≠Thinking High · tools | +16.0 pts#19 vs #78 of 97 |
| MedScribeClinical Workflows · Rubric pass rate (%) | 91.4Max · tools | 67.7≠Thinking High · tools | +23.7 pts#1 vs #90 of 99 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 28.8Thinking High · no tools | 2.8≠Thinking · no tools | +26.0 pts#10 vs #81 of 96 |
| MMMU-ProVisual Understanding · Accuracy | 89.8Max · tools | 64.9≠Thinking · no tools | +24.9 pts#2 vs #93 of 136 |
| AA-AnalystAgentData Analysis · pass^5 (%) | 56.3Max · tools + web | 12.5≠Normal · tools + web | +43.8 pts#5 vs #34 of 39 |
| AA-BriefcaseOffice & Business · AA-Briefcase Elo | 1800Max · tools | 519≠Normal · tools | 3.47×#2 vs #19 of 20 |
| EMB (Excel Modeling)Finance · Accuracy (%) | 75.9Max · tools | 11.2≠Thinking High · tools | +64.8 pts#4 vs #76 of 78 |
| Finance Agent v2Finance · Score | 58.6Max · tools | 32.1≠Thinking High · tools | 1.83×#10 vs #67 of 75 |
| Harvey's Legal Agent BenchmarkLegal · Accuracy (%) | 3.8Max · tools | 0.42≠Thinking High · tools | +3.3 pts#31 vs #43 of 52 |
| Legal Research BenchLegal · Score | 50.5Max · tools | 9.1≠Thinking High · tools | 5.53×#8 vs #83 of 91 |
| SAGEService Workflows · Accuracy (%) | 45.8Max · tools | 37.6≠Thinking High · tools | +8.2 pts#37 vs #62 of 82 |
| Tax Agent BenchFinance · Accuracy (%, rubric × citation quality) | 70.5Max · tools | 4.5≠Thinking High · tools | +66.0 pts#6 vs #73 of 74 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 59.1Max · task tools | 4.6≠Thinking High · tools | 12.7×#5 vs #59 of 62 |
| Vals IndexCapability IndicesNot counted · This metric has no better direction | 69.7Max · task tools | 17.9≠Thinking High · tools | — |
Maintenance & Optimization · Hidden-test pass rate (%)
Code Generation & Editing · Weighted rubric score (%)
Code Generation & Editing · Pass rate (%)
Mathematics · Accuracy (%)
Data Analysis · pass^5 (%)
Office & Business · AA-Briefcase Elo
Finance · Accuracy (%)
Finance · Score
Legal · Accuracy (%)
Legal · Score
Service Workflows · Accuracy (%)
Finance · Accuracy (%, rubric × citation quality)
Capability Indices · Index (0-100)
Capability Indices
Not counted · This metric has no better direction
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 · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
Claude Opus 5.5 costs65%morethan Mistral Medium 3.5 per million tokens.Using standard rates at this token mix.
Claude Opus 5.5: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-10-21; full 1M context at standard price
Mistral Medium 3.5: Standard pricing. Mistral official pricing verified 2026-11-13; Deprecated 2026-10-31, scheduled retirement 2027-04-30 per docs.mistral.ai/models
| Price / 1M tokens | Claude Opus 5.5 | Mistral Medium 3.5 | Ratio |
|---|---|---|---|
| Input | $4.00 | $1.50 | 2.67× |
| Output | $20.00 | $7.50 | 2.67× |
| Cached input | $0.20 | — | — |
| Cache write | — | — | — |
| Blended | $4.46 | $2.70 | 1.65× |
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 5.5 relative to Mistral Medium 3.5
Summary: 1 of 2 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 1 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 5.5 3.3× more expensive
Settings or sources differ
Claude Opus 5.5 3.3× more expensive
Settings or sources differ
Effort, protocol, aggregation and source match; other run conditions may differ.
Claude Opus 5.5
Mistral Medium 3.5
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Opus 5.5
Mistral Medium 3.5
| Benchmark | Claude Opus 5.5 | Mistral Medium 3.5 |
|---|---|---|
| AA Coding Agent Index v1.5 | Not reported | |
| AA Coding Agent Index v2.0 | Not reported | |
| AA Intelligence Index v4.3 | 2 more settings | Not reported |
| AA-Briefcase | 2 more settings | Not reported |
| AA-LCR | 2 more settings | Not reported |
| AA-Omniscience | 2 more settings | Not reported |
| Agent Arena | $5.538 / task Median · Net improvement index (percentage points vs. average model): 19.5 Max (With Tools + Internet) $2.1144 / task Median · Net improvement index (percentage points vs. average model): 14.2 Thinking High (With Tools + Internet) | Not reported |
| Agent Arena Index | $5.9958 / task Median · Agent Arena Index rating (index points): 1336 Max (With Tools + Internet) $1.9627 / task Median · Agent Arena Index rating (index points): 1273 Thinking High (With Tools + Internet) | Not reported |
| Agents' Last Exam v1.0 | Not reported | |
| ARC-AGI-1 | 2 more settings | Not reported |
| ARC-AGI-2 | 2 more settings$0.3369 / task Average · ARC-AGI-2 Score: 87.5 Thinking Medium (No Tools) | Not reported |
| AutomationBench | $0.71 / task Average · Pass Rate: 33.0 Thinking High (With Tools) · zapier-private-heldout $0.51 / task Average · Pass Rate: 24.2 Thinking Low (With Tools) · zapier-private-heldout 2 more settings$0.65 / task Average · Pass Rate: 29.5 Thinking Medium (With Tools) · zapier-private-heldout | Not reported |
| BenchCAD 2.1 Core | $2.571 / task Average · Core mean geometry/circuit score (0–100): 29.7 Thinking High (With Tools) $0.0312 / task Average · Core mean geometry/circuit score (0–100): 6.3 Thinking Low (With Tools) 2 more settings$1.114 / task Average · Core mean geometry/circuit score (0–100): 19.1 Thinking Medium (With Tools) $8.906 / task Average · Core mean geometry/circuit score (0–100): 54.5 Extra-High (With Tools) | Not reported |
| Chartography | 1 more settings | Not reported |
| CritPt | 1 more settings | Not reported |
| CursorBench 4.0 | $3.97 / task Average · Task score: 56.0 Thinking High (With Tools) · cursorbench-4-0-cursor-agent $1.17 / task Average · Task score: 43.7 Thinking Low (With Tools) · cursorbench-4-0-cursor-agent 2 more settings$2.91 / task Average · Task score: 52.5 Thinking Medium (With Tools) · cursorbench-4-0-cursor-agent $6.98 / task Average · Task score: 56.0 Extra-High (With Tools) · cursorbench-4-0-cursor-agent | Not reported |
| CWE-bench v1 | Not reported | |
| FrontierCode 1.1 | $6.191 / task Average · Weighted rubric score: 54.4 Max (With Tools) · frontiercode-1-1-main $1.0896 / task Average · Weighted rubric score: 54.0 Thinking High (With Tools) · frontiercode-1-1-main $0.4036 / task Average · Weighted rubric score: 47.3 Thinking Low (With Tools) · frontiercode-1-1-main 2 more settings$0.8016 / task Average · Weighted rubric score: 54.6 Thinking Medium (With Tools) · frontiercode-1-1-main $2.2545 / task Average · Weighted rubric score: 51.4 Extra-High (With Tools) · frontiercode-1-1-main | Not reported |
| FrontierCode 2.0 | $2.423 / task Average · Weighted rubric score: 34.5 Thinking High (With Tools + Internet) 2 more settings$1.699 / task Average · Weighted rubric score: 30.1 Thinking Medium (With Tools + Internet) | Not reported |
| FrontierCS 2 | Not reported | |
| FrontierSWE v2 | Not reported | |
| GDP.pdf | 1 more settings | Not reported |
| GDPval-AA v2 | 2 more settings | Not reported |
| GDPx | Not reported | |
| HLE | $0.0619 / task Average · Accuracy: 54.7 Thinking Medium (No Tools) · Text only · text-only 6 more settings | Not reported |
| LiveBench 2.0 | $3.6255 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 86.5 Max (task-specific tools) | Not reported |
| SciCode | 1 more settings | Not reported |
| SWE-sweep | Not reported | |
| Terminal-Bench 2.1 | Not reported | |
| Terminal-Bench 4.0 | 5 more settings$4.6958 / task Average · Resolution rate (%): 67.7 Thinking High (With Tools) · vals-effort-study $2.5888 / task Average · Resolution rate (%): 40.4 Thinking Low (With Tools) · vals-effort-study $13.20 / task Average · Resolution rate (%): 65.2 Unspecified Reasoning (With Tools) · vals-mini-swe | Not reported |
| Terminal-Bench 4.1 | Not reported | |
| Terminal-Bench-Science 0.1 | Not reported | |
| Vals Index v2.1 | Not reported | |
| Vals Index v2.2 | Not reported | |
| Vibe Code Bench 1-100 | Not reported | |
| WANDR | 2 more settings | Not reported |
| WeirdML v3 | $10.2494 / task Average · Official score (0-1; higher is better): 0.31 Extra-High (With Tools) | Not reported |
| Attribute | Claude Opus 5.5 | Mistral Medium 3.5 |
|---|---|---|
| Availability | Available | Available |
| Provider | Anthropic | MistralAI |
| Model type | Reasoning model | Chat model |
| First release | September 22, 2026 | May 1, 2026 |
| Context window | 1M tokens | 256K tokens |
| Max output | 128K tokens | — |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual, Coding, Function calling, Tool use, Computer use, Long context | Coding, Tool use, Function calling, Long context |
| Knowledge cutoff | 2026-06 | — |
| Input modalities | Text, Image | Text |
| 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 5.5 or Mistral Medium 3.5? 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.