Across 4 comparable benchmarks, GPT-6.2 Sol has better reported results on 4 and worse on 0.
Moderate evidence4 of 4 shared results countedHow we compare
Price basis: Standard rates vs listed rates. The price section below shows the applicable periods.
| Fact | Mistral Large 4 | GPT-6.2 Sol |
|---|---|---|
| Context window | 1M tokens | 1.05M tokens |
| Input / 1M | $0.700 | $2.00 |
| Output / 1M | $2.10 | $10.00 |
| Blended (4:1) | $0.526 | $2.23 |
| Status | Preview | Available |
| Released | October 6, 2026 | November 17, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 2 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
GPT-6.2 Sol ahead on 3 of 3 counted
No shared comparable results for Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Mistral Large 4 relative to GPT-6.2 Sol. Reading the results
| Benchmark | Mistral Large 4 | GPT-6.2 Sol | Gap |
|---|---|---|---|
| FrontierCode 2.0Repository Engineering · Weighted rubric score | 8.8Thinking High · tools + web | 44.4≠Max · tools | 0.20×#30 vs #5 of 36 |
| Terminal-Bench 4.0Agentic Development · Resolution rate (%) | 28.3Thinking · tools | 60.1≠Max · tools | −31.8 pts#40 vs #10 of 90 |
| Terminal-Bench 4.1Agentic Development · Resolution rate (%) | 27.6Max · tools | 61.5Max · tools | −33.9 pts#26 vs #6 of 48 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 25.3Thinking · tools | 57.3≠Max · task tools | 0.44×#39 vs #6 of 62 |
Repository Engineering · Weighted rubric score
Agentic Development · Resolution rate (%)
Agentic Development · Resolution rate (%)
Capability Indices · Index (0-100)
A dash means no result is available.
| Benchmark | Mistral Large 4 | GPT-6.2 Sol |
|---|---|---|
| FrontierMath Tier 4 v2 | 26.8 | — |
| FrontierMath v2 | 58.6 | — |
| GPQA Diamond | — | 96.0 |
| HLE-Diamond | — | 64.0 |
| MysteryMechanism | 12.2 | — |
| ProofBench v1.1 (Lean 4) | 10.0 | — |
| Terminal-Bench-Science 0.1 | — | 67.6 |
| AA Cyber Index | 49.5 | — |
| Code Migration | 30.6 | — |
| DeepSWE | 61.7 | — |
| FrontierCode 1.1 | 39.4 | — |
| IOI (Vals v2) | 45.3 | — |
| SWE-Atlas-QnA | 59.4 | — |
| SWE-sweep | — | 11.6 |
| Vibe Code Bench 1-100 | 14.3 | — |
| Vibe Code Bench v1.1 | 78.4 | — |
| WeirdML v3 | 0.044 | — |
| SEC-Bench Pro | — | 88.3 |
| Agent Arena Index | — | 1241 |
| LiveBench | 71.8 | — |
| LiveBench 2.0 | 47.7 | — |
| Vals Index v2.1 | 48.0 | — |
| Vals Index v2.2 | — | 66.1 |
| Chartography | — | 75.1 |
| GDP.pdf | — | 30.0 |
| LVBench | — | 87.6 |
| MMMU-Pro | 76.4 | — |
| AA-AnalystAgent | — | 53.8 |
| AutomationBench | — | 41.3 |
| BenchCAD 2.1 Core | — | 73.1 |
| EMB (Excel Modeling) | 56.2 | — |
| Finance Agent v2 | 54.7 | — |
| Harvey LAB-AA v1.1 (All-pass) | — | 11.1 |
| Harvey's Legal Agent Benchmark | 15.8 | — |
| Legal Research Bench | 31.7 | — |
| Tax Agent Bench | 63.3 | — |
| SimpleBench | — | 69.9 |
| BioMysteryBench | 67.0 | — |
| HealthBench Professional | — | 65.9 |
| MedCode | 40.7 | — |
| MedScribe | 80.4 | — |
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%
Mistral Large 4 costs76%lessthan GPT-6.2 Sol per million tokens — 4.2× cheaper.Using standard rates vs listed rates at this token mix.
Mistral Large 4: Standard pricing. Mistral official pricing (docs.mistral.ai/inference/pricing) verified 2026-11-13; price reduced from $1.36 Mistral official pricing (docs.mistral.ai/inference/pricing) verified 2026-11-13; price reduced from $4.18
GPT-6.2 Sol: Standard pricing. context_length <= 272000 Official OpenAI per-million-token pricing. Full request uses the applicable input-token band. context_length > 272000
| Price / 1M tokens | Mistral Large 4 | GPT-6.2 Sol | Ratio |
|---|---|---|---|
| Input | $0.70 | $2.00 | 0.35× |
| Output | $2.10 | $10.00 | 0.21× |
| Cached input | $0.07 | $0.10 | 0.70× |
| Cache write | — | $2.50 | — |
| Blended | $0.526 | $2.23 | 0.24× |
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 Large 4 relative to GPT-6.2 Sol
Summary: 2 of 2 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 2 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 Large 4 3.5× more expensive
Mistral Large 4 1.7× more expensive
Mistral Large 4 3.5× more expensive
Mistral Large 4 1.7× more expensive
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Large 4
GPT-6.2 Sol
Effort, protocol, aggregation and source match; other run conditions may differ.
Mistral Large 4
GPT-6.2 Sol
| Benchmark | Mistral Large 4 | GPT-6.2 Sol |
|---|---|---|
| Agent Arena Index | Not reported | $0.9817 / task Median · Agent Arena Index rating (index points): 1241 Max (With Tools + Internet) |
| BenchCAD 2.1 Core | Not reported | $0.75 / task Average · Core mean geometry/circuit score (0–100): 64.0 Thinking High (With Tools) $0.15 / task Average · Core mean geometry/circuit score (0–100): 23.9 Thinking Low (With Tools) 2 more settings$0.29 / task Average · Core mean geometry/circuit score (0–100): 41.2 Thinking Medium (With Tools) $1.00 / task Average · Core mean geometry/circuit score (0–100): 70.4 Extra-High (With Tools) |
| BioMysteryBench | Not reported | |
| Code Migration | Not reported | |
| EMB (Excel Modeling) | Not reported | |
| Finance Agent v2 | Not reported | |
| Harvey's Legal Agent Benchmark | Not reported | |
| IOI (Vals v2) | Not reported | |
| Legal Research Bench | Not reported | |
| LiveBench 2.0 | $0.6726 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 47.7 Thinking High (With Tools) | Not reported |
| MedCode | Not reported | |
| MedScribe | Not reported | |
| MysteryMechanism | Not reported | |
| ProofBench v1.1 (Lean 4) | Not reported | |
| Tax Agent Bench | $2.7212 / task Average · Accuracy (%, rubric × citation quality): 63.3 Thinking High (With Tools) | Not reported |
| Terminal-Bench 4.0 | $2.73 / task Average · Resolution rate (%): 26.3 Unspecified Reasoning (With Tools) · vals-mini-swe | Not reported |
| Terminal-Bench 4.1 | Not reported | |
| Vals Index v2.1 | Not reported | |
| Vibe Code Bench 1-100 | Not reported | |
| Vibe Code Bench v1.1 | Not reported | |
| WeirdML v3 | $28.7748 / task Average · Official score (0-1; higher is better): 0.044 Thinking High (With Tools) | Not reported |
| Attribute | Mistral Large 4 | GPT-6.2 Sol |
|---|---|---|
| Availability | Preview | Available |
| Provider | MistralAI | OpenAI |
| Model type | Multimodal model | Reasoning model |
| First release | October 6, 2026 | November 17, 2026 |
| Context window | 1M tokens | 1.05M tokens |
| Max output | — | 128K tokens |
| Total parameters | 1.1T | — |
| Active parameters | 49B | — |
| Architecture | Mixture of Experts (MoE) | Undisclosed |
| Documented capabilities | Reasoning, Multilingual, Function calling, Tool use, Structured output, Long context, Coding | Reasoning, Coding, Function calling, Tool use, Structured output, Computer use, Web search, Long context |
| Knowledge cutoff | — | 2026-06-01 |
| Input modalities | Text, Image | Text, Image |
| 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 Large 4 or GPT-6.2 Sol? 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.