Across 21 comparable benchmarks, Gemini 3.1 Pro Preview has better reported results on 19 and worse on 2. 1 further shared result is listed below, each with the reason it is not counted.
Strong evidence21 of 22 shared results countedHow we compare
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | Gemini 3.1 Pro Preview | Mistral Medium 3.5 |
|---|---|---|
| Context window | 1M tokens | 256K tokens |
| Input / 1M | $2.00 | $1.50 |
| Output / 1M | $12.00 | $7.50 |
| Blended (4:1) | $2.70 | $2.70 |
| Status | Preview | Available |
| Released | February 20, 2026 | May 1, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates β set your own mix.
Mean of 16 shared benchmarks grouped by capability on a 0β100 scale. How averages are built
Gemini 3.1 Pro Preview ahead on 3 of 3 counted
Gemini 3.1 Pro Preview ahead on 1 of 1 counted
Gemini 3.1 Pro Preview ahead on 3 of 3 counted
Gemini 3.1 Pro Preview ahead on 1 of 1 counted
Gemini 3.1 Pro Preview ahead on 3 of 3 counted
Gemini 3.1 Pro Preview ahead on 7 of 9 counted
No shared comparable results for Abstract reasoning, Computer & tool use, Research & search, Cybersecurity.
Published scores, with matching modes where available. Gaps show Gemini 3.1 Pro Preview relative to Mistral Medium 3.5. Reading the results
| Benchmark | Gemini 3.1 Pro Preview | Mistral Medium 3.5 | Gap |
|---|---|---|---|
| Code MigrationMaintenance & Optimization Β· Hidden-test pass rate (%) | 17.3Thinking High Β· tools | 5.1Thinking High Β· tools | +12.2 pts#55 vs #72 of 79 |
| Terminal Bench HardAgentic Development Β· Accuracy | 53.8Thinking Β· tools | 33.3Thinking Β· tools | +20.5 pts#8 vs #58 of 157 |
| Vibe Code Bench v1.1Code Generation & Editing Β· Pass rate (%) | 32.0Thinking High Β· tools | 2.9Thinking High Β· tools | +29.1 pts#77 vs #115 of 121 |
| ProofBench v1.1 (Lean 4)Mathematics Β· Accuracy (%) | 26.0Thinking High Β· tools | 9.0Thinking High Β· tools | +17.0 pts#40 vs #50 of 54 |
| SciCodeScientific Computing Β· Score | 58.7Thinking Β· no tools | 40.2Thinking Β· no tools | 1.46Γ#18 vs #75 of 98 |
| MedCodeClinical Workflows Β· Accuracy (%) | 59.1Thinking High Β· tools | 33.8Thinking High Β· tools | +25.3 pts#2 vs #78 of 97 |
| MedScribeClinical Workflows Β· Rubric pass rate (%) | 76.1Thinking High Β· tools | 67.7Thinking High Β· tools | +8.4 pts#70 vs #90 of 99 |
| Context ArenaMemory & Persistence Β· Accuracy (8 needles, 4K-128K context) | 84.7Thinking High Β· no tools | 32.0Thinking High Β· no tools | +52.7 pts#15 vs #58 of 61 |
| GDP.pdfDocuments & Charts Β· Task pass rate (%) | 17.8Thinking Β· no tools | 2.8Thinking Β· no tools | +15.0 pts#33 vs #81 of 96 |
| GeoBench ACWVisual Understanding Β· ACW country accuracy (%) | 84.0Normal Β· no tools | 56.0Normal Β· no tools | +28.0 pts#7 vs #29 of 32 |
| MMMU-ProVisual Understanding Β· Accuracy | 82.4Thinking Β· no tools | 64.9Thinking Β· no tools | +17.5 pts#19 vs #93 of 136 |
| AA-AnalystAgentData Analysis Β· pass^5 (%) | 41.3Thinking Β· tools + web | 12.5β Normal Β· tools + web | +28.8 pts#22 vs #34 of 39 |
| EMB (Excel Modeling)Finance Β· Accuracy (%) | 52.6Thinking High Β· tools | 11.2Thinking High Β· tools | +41.4 pts#56 vs #76 of 78 |
| Finance Agent v2Finance Β· Score | 43.0Thinking High Β· tools | 32.1Thinking High Β· tools | 1.34Γ#57 vs #67 of 75 |
| Harvey Lab-AALegal Β· Criterion pass rate (some entries report all-pass rate) | 58.9Thinking Β· tools | 69.1Thinking Β· tools | β10.2 pts#33 vs #29 of 39 |
| Harvey's Legal Agent BenchmarkLegal Β· Accuracy (%) | 0.0Thinking High Β· tools | 0.42Thinking High Β· tools | β0.4 pts#44 vs #43 of 52 |
| Legal Research BenchLegal Β· Score | 20.7Thinking High Β· tools | 9.1Thinking High Β· tools | 2.26Γ#68 vs #83 of 91 |
| SAGEService Workflows Β· Accuracy (%) | 48.7Thinking High Β· tools | 37.6Thinking High Β· tools | +11.1 pts#26 vs #62 of 82 |
| Tax Agent BenchFinance Β· Accuracy (%, rubric Γ citation quality) | 9.3Thinking High Β· tools | 4.5Thinking High Β· tools | +4.8 pts#66 vs #73 of 74 |
| ΟΒ³-BankingService Workflows Β· Score | 26.0Thinking High Β· tools | 15.1β Thinking Β· tools | 1.72Γ#46 vs #74 of 120 |
| ECICapability Indices Β· ECI score (capability index, higher is better) | 155Reported best | 141Reported best | 1.10Γ#31 vs #92 of 168 |
| Vals IndexCapability IndicesNot counted Β· This metric has no better direction | 41.9Thinking High Β· tools | 17.9Thinking High Β· tools | β |
Maintenance & Optimization Β· Hidden-test pass rate (%)
Agentic Development Β· Accuracy
Code Generation & Editing Β· Pass rate (%)
Mathematics Β· Accuracy (%)
Memory & Persistence Β· Accuracy (8 needles, 4K-128K context)
Documents & Charts Β· Task pass rate (%)
Visual Understanding Β· ACW country accuracy (%)
Visual Understanding Β· Accuracy
Data Analysis Β· pass^5 (%)
Finance Β· Accuracy (%)
Finance Β· Score
Legal Β· Criterion pass rate (some entries report all-pass rate)
Legal Β· Accuracy (%)
Legal Β· Score
Service Workflows Β· Accuracy (%)
Finance Β· Accuracy (%, rubric Γ citation quality)
Service Workflows Β· Score
Capability Indices Β· ECI score (capability index, higher is better)
Capability Indices
Not counted Β· 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 Β· 5% steps
Blended price = effective input rate at 90% cache hits Γ 80% + output rate Γ 20%
Gemini 3.1 Pro Preview and Mistral Medium 3.5 costabout the sameat this token mix.Using listed rates vs standard rates at this token mix.
Gemini 3.1 Pro Preview: Standard pricing. context_length <= 200K Google Gemini API official pricing (ai.google.dev/gemini-api/docs/pricing) verified 2026-10-21; paid tier; standard tier; prompts <= 200k tokens context_length > 200K Google Gemini API official pricing (ai.google.dev/gemini-api/docs/pricing) verified 2026-10-21; paid tier; standard tier; prompts > 200k tokens
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 | Gemini 3.1 Pro Preview | Mistral Medium 3.5 | Ratio |
|---|---|---|---|
| Input | $2.00 | $1.50 | 1.33Γ |
| Output | $12.00 | $7.50 | 1.60Γ |
| Cached input | $0.20 | β | β |
| Cache write | β | β | β |
| Blended | $2.70 | $2.70 | 1.00Γ |
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.
Gemini 3.1 Pro Preview relative to Mistral Medium 3.5
Summary: 1 of 1 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.
Gemini 3.1 Pro Preview 4.8Γ cheaper
Gemini 3.1 Pro Preview 4.8Γ cheaper
Effort, protocol, aggregation and source match; other run conditions may differ.
Gemini 3.1 Pro Preview
Mistral Medium 3.5
| Benchmark | Gemini 3.1 Pro Preview | Mistral Medium 3.5 |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| Agents' Last Exam v1.0 | $14.2116 / task Average Β· Pass rate: 16.4 Unspecified Reasoning (With Tools) Β· pass-rate | Not reported |
| AIME 2026 | Not reported | |
| AutomationBench | $0.30 / task Average Β· Pass Rate: 8.7 Thinking High (With Tools) Β· zapier-private-heldout $0.28 / task Average Β· Pass Rate: 4.6 Thinking Low (With Tools) Β· zapier-private-heldout $0.25 / task Average Β· Pass Rate: 7.0 Thinking Medium (With Tools) Β· zapier-private-heldout | Not reported |
| DeepSWE | Not reported | |
| IOI (Vals v2) | Not reported | |
| LiveBench 2.0 | $0.6982 / task Average Β· Overall score (equal-weight mean of 7 suite scores, 0-100): 54.9 Thinking High (With Tools) | Not reported |
| MysteryMechanism | Not reported | |
| Vals Index v2.1 | Not reported | |
| Vibe Code Bench 1-100 | Not reported |
| Attribute | Gemini 3.1 Pro Preview | Mistral Medium 3.5 |
|---|---|---|
| Availability | Preview | Available |
| Provider | Google Deep Mind | MistralAI |
| Model type | Multimodal model | Chat model |
| First release | February 20, 2026 | May 1, 2026 |
| Context window | 1M tokens | 256K tokens |
| Max output | 64K tokens | β |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual | Coding, Tool use, Function calling, Long context |
| Knowledge cutoff | 2025-01 | β |
| Input modalities | Text, Image, Audio, Video | 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 Gemini 3.1 Pro Preview 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.