Across 18 comparable benchmarks, Mistral Medium 3.5 has better reported results on 11 and worse on 7. 1 further shared result is listed below, each with the reason it is not counted.
Strong evidence18 of 19 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Gemini 3.1 Flash-Lite | Mistral Medium 3.5 |
|---|---|---|
| Context window | 1M tokens | 256K tokens |
| Input / 1M | $0.250 | $1.50 |
| Output / 1M | $1.50 | $7.50 |
| Blended (4:1) | $0.338 | $2.70 |
| Status | Deprecated | Available |
| Released | May 7, 2026 | May 1, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 12 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Mistral Medium 3.5 ahead on 3 of 3 counted
Gemini 3.1 Flash-Lite ahead on 2 of 3 counted
Gemini 3.1 Flash-Lite ahead on 2 of 2 counted
Mistral Medium 3.5 ahead on 7 of 8 counted
No shared comparable results for Abstract reasoning, Mathematics, Computer & tool use, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show Gemini 3.1 Flash-Lite relative to Mistral Medium 3.5. Reading the results
| Benchmark | Gemini 3.1 Flash-Lite | Mistral Medium 3.5 | Gap |
|---|---|---|---|
| Code MigrationMaintenance & Optimization · Hidden-test pass rate (%) | 4.6Thinking High · tools | 5.1Thinking High · tools | −0.5 pts#74 vs #72 of 79 |
| Terminal Bench HardAgentic Development · Accuracy | 24.2Thinking · tools | 33.3Thinking · tools | −9.1 pts#83 vs #58 of 157 |
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 0.0Thinking High · tools | 2.9Thinking High · tools | −2.9 pts#118 vs #115 of 121 |
| SciCodeScientific Computing · Score | 43.4Thinking · no tools | 40.2Thinking · no tools | 1.08×#69 vs #75 of 98 |
| MedCodeClinical Workflows · Accuracy (%) | 47.6Thinking High · tools | 33.8Thinking High · tools | +13.9 pts#31 vs #78 of 97 |
| MedScribeClinical Workflows · Rubric pass rate (%) | 63.9Thinking High · tools | 67.7Thinking High · tools | −3.8 pts#92 vs #90 of 99 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 7.8Thinking · no tools | 2.8Thinking · no tools | +5.0 pts#67 vs #81 of 96 |
| MMMU-ProVisual Understanding · Accuracy | 75.5Thinking · no tools | 64.9Thinking · no tools | +10.6 pts#62 vs #93 of 136 |
| AA-AnalystAgentData Analysis · pass^5 (%) | 8.8Thinking · tools + web | 12.5≠Normal · tools + web | −3.8 pts#37 vs #34 of 39 |
| EMB (Excel Modeling)Finance · Accuracy (%) | 8.6Thinking High · tools | 11.2Thinking High · tools | −2.6 pts#78 vs #76 of 78 |
| Finance Agent v2Finance · Score | 30.0Thinking High · tools | 32.1Thinking High · tools | 0.93×#69 vs #67 of 75 |
| Harvey Lab-AALegal · Criterion pass rate (some entries report all-pass rate) | 31.1Thinking · tools | 69.1Thinking · tools | −37.9 pts#36 vs #29 of 39 |
| Harvey's Legal Agent BenchmarkLegal · Accuracy (%) | 0.0Thinking High · tools | 0.42Thinking High · tools | −0.4 pts#44 vs #43 of 52 |
| Legal Research BenchLegal · Score | 3.4Thinking High · tools | 9.1Thinking High · tools | 0.37×#87 vs #83 of 91 |
| SAGEService Workflows · Accuracy (%) | 49.5Thinking High · tools | 37.6Thinking High · tools | +11.9 pts#20 vs #62 of 82 |
| τ³-BankingService Workflows · Score | 9.7Thinking · tools | 15.1Thinking · tools | 0.64×#94 vs #74 of 120 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 6.2Thinking High · tools | 4.6Thinking High · tools | 1.33×#57 vs #59 of 62 |
| ECICapability Indices · ECI score (capability index, higher is better) | 144Reported best | 141Reported best | 1.02×#81 vs #92 of 168 |
| Vals IndexCapability IndicesNot counted · This metric has no better direction | 15.5Thinking High · tools | 17.9Thinking High · tools | — |
Maintenance & Optimization · Hidden-test pass rate (%)
Agentic Development · Accuracy
Code Generation & Editing · Pass rate (%)
Data Analysis · pass^5 (%)
Finance · Accuracy (%)
Finance · Score
Legal · Criterion pass rate (some entries report all-pass rate)
Legal · Accuracy (%)
Legal · Score
Service Workflows · Accuracy (%)
Service Workflows · Score
Capability Indices · Index (0-100)
Capability Indices · ECI score (capability index, higher is better)
Capability Indices
Not counted · This metric has no better direction
A dash means no result is available.
| Benchmark | Gemini 3.1 Flash-Lite | Mistral Medium 3.5 |
|---|---|---|
| CritPt | 1.1 | — |
| FrontierMath v2 | 27.7 | — |
| GPQA Diamond | 81.8 | — |
| HLE | 8.6 | — |
| ProofBench v1.1 (Lean 4) | — | 9.0 |
| FrontierCode 1.1 | — | 8.0 |
| Terminal-Bench 4.0 | 0.5 | — |
| MCP-Atlas | 57.1 | — |
| PinchBench v2 | 80.5 | — |
| AA Intelligence Index (historical versions) | — | 30.4 |
| LiveBench | 61.7 | — |
| Context Arena | — | 32.0 |
| GeoBench ACW | — | 56.0 |
| AA-Briefcase | — | 519 |
| Tax Agent Bench | — | 4.5 |
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%
Gemini 3.1 Flash-Lite costs87%lessthan Mistral Medium 3.5 per million tokens — 8× cheaper.Using standard rates at this token mix.
Gemini 3.1 Flash-Lite: Standard pricing. Google Gemini API official pricing (ai.google.dev/gemini-api/docs/pricing) verified 2026-10-21; paid tier; standard tier
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 Flash-Lite | Mistral Medium 3.5 | Ratio |
|---|---|---|---|
| Input | $0.25 | $1.50 | 0.17× |
| Output | $1.50 | $7.50 | 0.20× |
| Cached input | $0.025 | — | — |
| Cache write | — | — | — |
| Blended | $0.338 | $2.70 | 0.13× |
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 Flash-Lite 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 Flash-Lite 5.3× cheaper
Gemini 3.1 Flash-Lite 5.3× cheaper
Effort, protocol, aggregation and source match; other run conditions may differ.
Gemini 3.1 Flash-Lite
Mistral Medium 3.5
| Benchmark | Gemini 3.1 Flash-Lite | Mistral Medium 3.5 |
|---|---|---|
| Tax Agent Bench | Not reported | $1.4437 / task Average · Accuracy (%, rubric × citation quality): 4.5 Thinking High (With Tools) |
| Attribute | Gemini 3.1 Flash-Lite | Mistral Medium 3.5 |
|---|---|---|
| Availability | Deprecated | Available |
| Provider | Google Deep Mind | MistralAI |
| Model type | Multimodal model | Chat model |
| First release | May 7, 2026 | May 1, 2026 |
| Context window | 1M tokens | 256K tokens |
| Max output | 66K tokens | — |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Multilingual, Tool use, Long context | Coding, Tool use, Function calling, Long context |
| Knowledge cutoff | 2025-01 | — |
| 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 Flash-Lite 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.