Across 1 comparable benchmark, Kimi K3.1 Pro has better reported results on 1 and worse on 0.
Limited evidence1 of 1 shared result countedHow we compare
Mistral Medium 3.1 is no longer available through its official API. API prices for this version are not shown.
Price basis: Listed rates vs international rates. The price section below shows the applicable periods.
| Fact | Mistral Medium 3.1 | Kimi K3.1 Pro |
|---|---|---|
| Context window | 128K tokens | 1M tokens |
| Input / 1M | — | $2.00 |
| Output / 1M | — | $10.00 |
| Blended (4:1) | — | $2.30 |
| Status | Retired | Available |
| Released | August 12, 2025 | October 24, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
No shared benchmark here has a fixed 0–100 scale, so only the counted results are shown. How averages are built
Kimi K3.1 Pro ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Coding, Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show Mistral Medium 3.1 relative to Kimi K3.1 Pro. Reading the results
| Benchmark | Mistral Medium 3.1 | Kimi K3.1 Pro | Gap |
|---|---|---|---|
| Creative WritingWriting | 1476Normal · no tools | 2173Normal · no tools | 0.68×#64 vs #2 of 121 |
Writing
A dash means no result is available.
| Benchmark | Mistral Medium 3.1 | Kimi K3.1 Pro |
|---|---|---|
| GPQA Diamond | — | 94.2 |
| HLE | — | 70.0 |
| HLE-Diamond | — | 39.4 |
| MysteryMechanism | — | 26.1 |
| ProofBench v1.1 (Lean 4) | — | 91.0 |
| SciCode | — | 60.8 |
| Terminal-Bench-Science 0.1 | — | 38.1 |
| AA Cyber Index | — | 64.1 |
| DeepSWE | — | 75.5 |
| FrontierCS 2 | — | 31.3 |
| IOI (Vals v2) | — | 81.9 |
| Kimi Code Bench 2.0 | — | 77.1 |
| ProgramBench | — | 31.5 |
| SWE-Marathon | — | 51.7 |
| Terminal-Bench 4.0 | — | 65.7 |
| Terminal-Bench 4.1 | — | 59.7 |
| Vibe Code Bench v1.1 | — | 84.9 |
| WeirdML v3 | — | 0.099 |
| Agents' Last Exam v1.0 | — | 43.2 |
| Agents' Last Exam v1.1 | — | 43.4 |
| APEX-Agents | — | 67.2 |
| AutomationBench-AA | — | 73.3 |
| OSWorld 2.1 | — | 79.6 |
| AA Intelligence Index v4.3 | — | 55.1 |
| AA Intelligence Index v5.0.1 | — | 46.5 |
| Agent Arena | — | 4.8 |
| Agent Arena Index | — | 1145 |
| LiveBench 2.0 | — | 76.7 |
| Vals Index v2.2 | — | 62.5 |
| AA-LCR | — | 86.3 |
| BabyVision | — | 95.4 |
| Chartography | — | 90.2 |
| CharXiv RQ | — | 93.4 |
| GDP.pdf | — | 34.6 |
| MMMU-Pro | — | 90.0 |
| PerceptionBench | — | 68.3 |
| ZeroBench Main | — | 63.0 |
| AA-Briefcase | — | 1678 |
| BenchCAD | — | 92.7 |
| GDPval-AA v2 | — | 1846 |
| Legal Research Bench | — | 37.5 |
| Office QA Pro | — | 67.4 |
| Tax Agent Bench | — | 24.9 |
| BrowseComp | — | 92.1 |
| BioMysteryBench | — | 75.6 |
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%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
Mistral Medium 3.1: Listed text rates; pricing tier not confirmed.
Kimi K3.1 Pro: International pricing. International API; kimi-k3.1-pro; https://platform.moonshot.ai/docs/pricing/chat
| Price / 1M tokens | Mistral Medium 3.1 | Kimi K3.1 Pro | Ratio |
|---|---|---|---|
| Input | — | $2.00 | — |
| Output | — | $10.00 | — |
| Cached input | — | $0.20 | — |
| Cache write | — | — | — |
| Blended | — | $2.30 | — |
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.
| Benchmark | Mistral Medium 3.1 | Kimi K3.1 Pro |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| AA-Briefcase | Not reported | |
| AA-LCR | Not reported | |
| Agent Arena | Not reported | $1.173 / task Median · Net improvement index (percentage points vs. average model): 4.8 Max (With Tools + Internet) |
| Agent Arena Index | Not reported | $1.1433 / task Median · Agent Arena Index rating (index points): 1145 Max (With Tools + Internet) |
| BioMysteryBench | Not reported | |
| FrontierCS 2 | Not reported | |
| HLE | Not reported | |
| IOI (Vals v2) | Not reported | |
| Legal Research Bench | Not reported | |
| LiveBench 2.0 | Not reported | $0.7172 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 76.7 Max (task-specific tools) |
| MysteryMechanism | Not reported | |
| ProofBench v1.1 (Lean 4) | Not reported | |
| SciCode | Not reported | |
| Tax Agent Bench | Not reported | |
| Terminal-Bench 4.0 | Not reported | $4.57 / task Average · Resolution rate (%): 63.6 Unspecified Reasoning (With Tools) · vals-mini-swe |
| Terminal-Bench 4.1 | Not reported | |
| Vals Index v2.2 | Not reported | |
| Vibe Code Bench v1.1 | Not reported | |
| WeirdML v3 | Not reported | $10.0245 / task Average · Official score (0-1; higher is better): 0.099 Extra-High (With Tools) |
| Attribute | Mistral Medium 3.1 | Kimi K3.1 Pro |
|---|---|---|
| Availability | Retired | Available |
| Provider | MistralAI | Moonshot AI |
| Model type | Chat model | Reasoning model |
| First release | August 12, 2025 | October 24, 2026 |
| Context window | 128K tokens | 1M tokens |
| Max output | — | 1M tokens |
| Total parameters | — | 2.8T |
| Active parameters | — | 104B |
| Architecture | Undisclosed | Mixture of Experts (MoE) |
| Documented capabilities | — | Reasoning, Multilingual, Coding, Function calling, Tool use, Structured output, Long context |
| Input modalities | Text | Text, Image, Video |
| Output modalities | Text | Text |
| Weights | Closed / API only | Open weights |
| Licence | — | Kimi K3 License |
| Commercial use | Proprietary | Commercial use allowed, with conditions |
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 3.1 or Kimi K3.1 Pro? 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.