Across 5 comparable benchmarks, Qwen4-Max has better reported results on 5 and worse on 0.
Moderate evidence5 of 5 shared results countedHow we compare
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | Qwen4-Max | Mistral Medium 3.5 |
|---|---|---|
| Context window | 1M tokens | 256K tokens |
| Input / 1M | $1.50 | $1.50 |
| Output / 1M | $4.50 | $7.50 |
| Blended (4:1) | $1.07 | $2.70 |
| Status | Available | Available |
| Released | November 16, 2026 | May 1, 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
Qwen4-Max ahead on 1 of 1 counted
Qwen4-Max ahead on 3 of 3 counted
No shared comparable results for Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show Qwen4-Max relative to Mistral Medium 3.5. Reading the results
| Benchmark | Qwen4-Max | Mistral Medium 3.5 | Gap |
|---|---|---|---|
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 86.7Max · tools | 2.9≠Thinking High · tools | +83.8 pts#16 vs #115 of 121 |
| AA-BriefcaseOffice & Business · AA-Briefcase Elo | 1686Max · tools | 519≠Normal · tools | 3.25×#6 vs #19 of 20 |
| Legal Research BenchLegal · Score | 41.4Max · tools | 9.1≠Thinking High · tools | 4.53×#32 vs #83 of 91 |
| Tax Agent BenchFinance · Accuracy (%, rubric × citation quality) | 40.4Max · tools | 4.5≠Thinking High · tools | +35.9 pts#29 vs #73 of 74 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 42.9Max · task tools | 4.6≠Thinking High · tools | 9.23×#18 vs #59 of 62 |
Code Generation & Editing · Pass rate (%)
Office & Business · AA-Briefcase Elo
Legal · Score
Finance · Accuracy (%, rubric × citation quality)
Capability Indices · Index (0-100)
A dash means no result is available.
| Benchmark | Qwen4-Max | Mistral Medium 3.5 |
|---|---|---|
| HLE | 51.7 | — |
| ProofBench v1.1 (Lean 4) | — | 9.0 |
| SciCode | — | 40.2 |
| Code Migration | — | 5.1 |
| DeepSWE | 75.7 | — |
| FrontierCode 1.1 | — | 8.0 |
| ProgramBench | 31.5 | — |
| Terminal Bench Hard | — | 33.3 |
| Terminal-Bench 4.0 | 48.5 | — |
| Agents' Last Exam v1.1 | 46.7 | — |
| APEX-Agents | 72.1 | — |
| AutomationBench-AA | 72.5 | — |
| AA Intelligence Index (historical versions) | — | 30.4 |
| Agent Arena | 7.5 | — |
| Agent Arena Index | 1189 | — |
| Code Arena WebDev | 1750 | — |
| ECI | — | 141 |
| Vals Index | — | 17.9 |
| Vals Index v2.2 | 63.2 | — |
| Context Arena | — | 32.0 |
| Chartography | 54.2 | — |
| GDP.pdf | — | 2.8 |
| GeoBench ACW | — | 56.0 |
| MMMU-Pro | — | 64.9 |
| AA-AnalystAgent | — | 12.5 |
| EMB (Excel Modeling) | — | 11.2 |
| Finance Agent v2 | — | 32.1 |
| Harvey Lab-AA | — | 69.1 |
| Harvey's Legal Agent Benchmark | — | 0.42 |
| JobBench | 68.5 | — |
| SAGE | — | 37.6 |
| τ³-Banking | — | 15.1 |
| BioMysteryBench | 85.0 | — |
| MedCode | — | 33.8 |
| MedScribe | — | 67.7 |
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%
Qwen4-Max costs60%lessthan Mistral Medium 3.5 per million tokens — 2.5× cheaper.Using listed rates vs standard rates at this token mix.
Qwen4-Max: International pricing. context_length <= 1000000 QwenCloud International (Singapore)
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 | Qwen4-Max | Mistral Medium 3.5 | Ratio |
|---|---|---|---|
| Input | $1.50 | $1.50 | 1.00× |
| Output | $4.50 | $7.50 | 0.60× |
| Cached input | $0.075 | — | — |
| Cache write | $1.88 | — | — |
| Blended | $1.07 | $2.70 | 0.40× |
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.
Qwen4-Max relative to Mistral Medium 3.5
No matching recorded settings for a cost summary.
Summary: 0 of 2 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 0 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ or are not recorded; interpret the comparison with care.
Qwen4-Max
Mistral Medium 3.5
Settings or sources differ or are not recorded; interpret the comparison with care.
Qwen4-Max
Mistral Medium 3.5
| Benchmark | Qwen4-Max | Mistral Medium 3.5 |
|---|---|---|
| Agent Arena Index | $1.4079 / task Median · Agent Arena Index rating (index points): 1189 Extra-High (Tools + Internet) | Not reported |
| Legal Research Bench | Not reported | |
| Terminal-Bench 4.0 | $3.90 / task Average · Resolution rate (%): 48.5 Unspecified Reasoning (With Tools) · vals-mini-swe | Not reported |
| Vals Index v2.2 | Not reported | |
| Vibe Code Bench v1.1 | Not reported |
| Attribute | Qwen4-Max | Mistral Medium 3.5 |
|---|---|---|
| Availability | Available | Available |
| Provider | Alibaba | MistralAI |
| Model type | Reasoning model | Chat model |
| First release | November 16, 2026 | May 1, 2026 |
| Context window | 1M tokens | 256K tokens |
| Max output | 384K tokens | — |
| Total parameters | 2.4T | — |
| Active parameters | 71B | — |
| Architecture | Mixture of Experts (MoE) | Undisclosed |
| Documented capabilities | Reasoning, Multilingual, Coding, Function calling, Tool use, Structured output, Web search, Long context | Coding, Tool use, Function calling, Long context |
| Input modalities | Text, Image, 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 Qwen4-Max 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.