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: Standard rates vs listed rates. The price section below shows the applicable periods.
| Fact | Mistral Medium 3.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Context window | 256K tokens | 1M tokens |
| Input / 1M | $1.50 | $2.00 |
| Output / 1M | $7.50 | $12.00 |
| Blended (4:1) | $2.70 | $2.70 |
| Status | Available | Preview |
| Released | May 1, 2026 | February 20, 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 Mistral Medium 3.5 relative to Gemini 3.1 Pro Preview. Reading the results
| Benchmark | Mistral Medium 3.5 | Gemini 3.1 Pro Preview | Gap |
|---|---|---|---|
| Code MigrationMaintenance & Optimization Β· Hidden-test pass rate (%) | 5.1Thinking High Β· tools | 17.3Thinking High Β· tools | β12.2 pts#72 vs #55 of 79 |
| Terminal Bench HardAgentic Development Β· Accuracy | 33.3Thinking Β· tools | 53.8Thinking Β· tools | β20.5 pts#58 vs #8 of 157 |
| Vibe Code Bench v1.1Code Generation & Editing Β· Pass rate (%) | 2.9Thinking High Β· tools | 32.0Thinking High Β· tools | β29.1 pts#115 vs #77 of 121 |
| ProofBench v1.1 (Lean 4)Mathematics Β· Accuracy (%) | 9.0Thinking High Β· tools | 26.0Thinking High Β· tools | β17.0 pts#50 vs #40 of 54 |
| SciCodeScientific Computing Β· Score | 40.2Thinking Β· no tools | 58.7Thinking Β· no tools | 0.68Γ#75 vs #18 of 98 |
| MedCodeClinical Workflows Β· Accuracy (%) | 33.8Thinking High Β· tools | 59.1Thinking High Β· tools | β25.3 pts#78 vs #2 of 97 |
| MedScribeClinical Workflows Β· Rubric pass rate (%) | 67.7Thinking High Β· tools | 76.1Thinking High Β· tools | β8.4 pts#90 vs #70 of 99 |
| Context ArenaMemory & Persistence Β· Accuracy (8 needles, 4K-128K context) | 32.0Thinking High Β· no tools | 84.7Thinking High Β· no tools | β52.7 pts#58 vs #15 of 61 |
| GDP.pdfDocuments & Charts Β· Task pass rate (%) | 2.8Thinking Β· no tools | 17.8Thinking Β· no tools | β15.0 pts#81 vs #33 of 96 |
| GeoBench ACWVisual Understanding Β· ACW country accuracy (%) | 56.0Normal Β· no tools | 84.0Normal Β· no tools | β28.0 pts#29 vs #7 of 32 |
| MMMU-ProVisual Understanding Β· Accuracy | 64.9Thinking Β· no tools | 82.4Thinking Β· no tools | β17.5 pts#93 vs #19 of 136 |
| AA-AnalystAgentData Analysis Β· pass^5 (%) | 12.5Normal Β· tools + web | 41.3β Thinking Β· tools + web | β28.8 pts#34 vs #22 of 39 |
| EMB (Excel Modeling)Finance Β· Accuracy (%) | 11.2Thinking High Β· tools | 52.6Thinking High Β· tools | β41.4 pts#76 vs #56 of 78 |
| Finance Agent v2Finance Β· Score | 32.1Thinking High Β· tools | 43.0Thinking High Β· tools | 0.75Γ#67 vs #57 of 75 |
| Harvey Lab-AALegal Β· Criterion pass rate (some entries report all-pass rate) | 69.1Thinking Β· tools | 58.9Thinking Β· tools | +10.2 pts#29 vs #33 of 39 |
| Harvey's Legal Agent BenchmarkLegal Β· Accuracy (%) | 0.42Thinking High Β· tools | 0.0Thinking High Β· tools | +0.4 pts#43 vs #44 of 52 |
| Legal Research BenchLegal Β· Score | 9.1Thinking High Β· tools | 20.7Thinking High Β· tools | 0.44Γ#83 vs #68 of 91 |
| SAGEService Workflows Β· Accuracy (%) | 37.6Thinking High Β· tools | 48.7Thinking High Β· tools | β11.1 pts#62 vs #26 of 82 |
| Tax Agent BenchFinance Β· Accuracy (%, rubric Γ citation quality) | 4.5Thinking High Β· tools | 9.3Thinking High Β· tools | β4.8 pts#73 vs #66 of 74 |
| ΟΒ³-BankingService Workflows Β· Score | 15.1Thinking Β· tools | 26.0β Thinking High Β· tools | 0.58Γ#74 vs #46 of 120 |
| ECICapability Indices Β· ECI score (capability index, higher is better) | 141Reported best | 155Reported best | 0.91Γ#92 vs #31 of 168 |
| Vals IndexCapability IndicesNot counted Β· This metric has no better direction | 17.9Thinking High Β· tools | 41.9Thinking 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%
Mistral Medium 3.5 and Gemini 3.1 Pro Preview costabout the sameat this token mix.Using standard rates vs listed rates at this token mix.
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
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
| Price / 1M tokens | Mistral Medium 3.5 | Gemini 3.1 Pro Preview | Ratio |
|---|---|---|---|
| Input | $1.50 | $2.00 | 0.75Γ |
| Output | $7.50 | $12.00 | 0.63Γ |
| Cached input | β | $0.20 | 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.
Mistral Medium 3.5 relative to Gemini 3.1 Pro Preview
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.
Mistral Medium 3.5 4.8Γ more expensive
Mistral Medium 3.5 4.8Γ more expensive
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Medium 3.5
Gemini 3.1 Pro Preview
| Benchmark | Mistral Medium 3.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| Agents' Last Exam v1.0 | Not reported | $14.2116 / task Average Β· Pass rate: 16.4 Unspecified Reasoning (With Tools) Β· pass-rate |
| AIME 2026 | Not reported | |
| AutomationBench | Not reported | $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 |
| DeepSWE | Not reported | |
| IOI (Vals v2) | Not reported | |
| LiveBench 2.0 | Not reported | $0.6982 / task Average Β· Overall score (equal-weight mean of 7 suite scores, 0-100): 54.9 Thinking High (With Tools) |
| MysteryMechanism | Not reported | |
| Vals Index v2.1 | Not reported | |
| Vibe Code Bench 1-100 | Not reported |
| Attribute | Mistral Medium 3.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Availability | Available | Preview |
| Provider | MistralAI | Google Deep Mind |
| Model type | Chat model | Multimodal model |
| First release | May 1, 2026 | February 20, 2026 |
| Context window | 256K tokens | 1M tokens |
| Max output | β | 64K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Coding, Tool use, Function calling, Long context | Reasoning, Multilingual |
| Knowledge cutoff | β | 2025-01 |
| Input modalities | Text | Text, Image, Audio, Video |
| 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 Medium 3.5 or Gemini 3.1 Pro Preview? 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.