Across 1 comparable benchmark, Hy4 Flash has better reported results on 1 and worse on 0.
Limited evidence1 of 1 shared result countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Hy4 Flash | Mistral Medium 3.5 |
|---|---|---|
| Context window | 1M tokens | 256K tokens |
| Input / 1M | $0.070 | $1.50 |
| Output / 1M | $0.281 | $7.50 |
| Blended (4:1) | $0.064 | $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 1 shared benchmark grouped by capability on a 0–100 scale. How averages are built
Hy4 Flash ahead on 1 of 1 counted
No shared comparable results for Coding, Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Hy4 Flash relative to Mistral Medium 3.5. Reading the results
| Benchmark | Hy4 Flash | Mistral Medium 3.5 | Gap |
|---|---|---|---|
| MMMU-ProVisual Understanding · Accuracy | 75.8Thinking High · no tools | 64.9≠Thinking · no tools | +10.9 pts#60 vs #93 of 136 |
Visual Understanding · Accuracy
A dash means no result is available.
| Benchmark | Hy4 Flash | Mistral Medium 3.5 |
|---|---|---|
| HLE | 55.2 | — |
| ProofBench v1.1 (Lean 4) | — | 9.0 |
| SciCode | — | 40.2 |
| Code Migration | — | 5.1 |
| DeepSWE | 64.6 | — |
| FrontierCode 1.1 | — | 8.0 |
| ProgramBench | 16 | — |
| Terminal Bench Hard | — | 33.3 |
| Terminal-Bench 4.0 | 19.2 | — |
| Vibe Code Bench v1.1 | — | 2.9 |
| APEX-Agents | 63.1 | — |
| OSWorld 2.1 | 68.1 | — |
| AA Intelligence Index (historical versions) | — | 30.4 |
| AA Intelligence Index v5.0.1 | — | 4.6 |
| ECI | — | 141 |
| Vals Index | — | 17.9 |
| Context Arena | — | 32.0 |
| GDP.pdf | — | 2.8 |
| GeoBench ACW | — | 56.0 |
| ZeroBench Main | 45.8 | — |
| AA-AnalystAgent | — | 12.5 |
| AA-Briefcase | — | 519 |
| AutomationBench | 52.2 | — |
| EMB (Excel Modeling) | — | 11.2 |
| Finance Agent v2 | — | 32.1 |
| Harvey Lab-AA | — | 69.1 |
| Harvey's Legal Agent Benchmark | — | 0.42 |
| Legal Research Bench | — | 9.1 |
| SAGE | — | 37.6 |
| Tax Agent Bench | — | 4.5 |
| τ³-Banking | — | 15.1 |
| 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%
Hy4 Flash costs98%lessthan Mistral Medium 3.5 per million tokens — 42.2× cheaper.Using standard rates at this token mix.
Hy4 Flash: Standard pricing. Tencent official international list price per million tokens.
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 | Hy4 Flash | Mistral Medium 3.5 | Ratio |
|---|---|---|---|
| Input | $0.07 | $1.50 | 0.05× |
| Output | $0.281 | $7.50 | 0.04× |
| Cached input | $0.003 | — | — |
| Cache write | — | — | — |
| Blended | $0.064 | $2.70 | 0.02× |
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 | Hy4 Flash | Mistral Medium 3.5 |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| Tax Agent Bench | Not reported | $1.4437 / task Average · Accuracy (%, rubric × citation quality): 4.5 Thinking High (With Tools) |
| Attribute | Hy4 Flash | Mistral Medium 3.5 |
|---|---|---|
| Availability | Available | Available |
| Provider | Tencent AI Lab | MistralAI |
| Model type | Multimodal model | Chat model |
| First release | November 16, 2026 | May 1, 2026 |
| Context window | 1M tokens | 256K tokens |
| Total parameters | 224B | — |
| Active parameters | 12B | — |
| Architecture | Mixture of Experts (MoE) | Undisclosed |
| Documented capabilities | Multilingual, Long context, Tool use, Computer use, Web search | Coding, Tool use, Function calling, Long context |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Weights | Open weights announced · release pending | Closed / API only |
| Licence | Unspecified (weights release pending) | — |
| Commercial use | Not documented | 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 Hy4 Flash 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.