AI Models · Open-Source vs Open-Source

Hy3 Open vs Mistral Medium 3.5 Open

Hy3 vs Mistral Medium 3.5 compared — price per token, context window, multimodality, openness and which to choose. Full 2026 breakdown.

Prices & specs refreshed from live data · olud.ai

Open-model prices = cheapest provider via OpenRouter; official maker rates may be higher.

Choose Hy3 for the lower output price ($0.53/M vs $7.5/M). It also gives you the larger 262K context window.

Hy3 vs Mistral Medium 3.5 specs

SpecHy3Mistral Medium 3.5
MakerTencentMistral AI
TypeOpen-weightOpen-weight
Context window262K tokens262K tokens
Input price$0.13/M · free self-host$1.5/M · free self-host
Output price$0.53/M · free self-host$7.5/M · free self-host
Vision / multimodalNoYes
Tool / function callingYesYes
Self-hostableYesYes
LicenseOpen weightsApache 2.0

Feature comparison

CapabilityHy3Mistral Medium 3.5
Open weights (downloadable)✓✓
Self-hostable✓✓
Runs fully offline✓✓
Vision / multimodal✗✓
Tool / function calling✓✓
1M+ context window✗✗

Benchmarks: Hy3 vs Mistral Medium 3.5

Independent benchmark scores measured by Artificial Analysis. Higher is better (except latency).

Hy3 delivers 17× more intelligence per dollar.
Hy3
Mistral Medium 3.5
Intelligence index
41.2
29.9
Coding index
58.8
46.9
GPQA
89.7%
74.8%
Humanity's Last Exam
31.6%
12.8%
Long Context Reasoning
66.7%
61%
SciCode
47.6%
39.6%
τ-Bench Banking
20.8%
14.4%
Terminal-Bench
64.4%
50.6%
IFBench
–
68.8%
τ²-Bench
–
94.2%
Terminal-Bench Hard
–
33.3%
Speed
68.6 tok/s
59.8 tok/s
Latency
1.68s
0.77s
Intelligence per $
171
10

Benchmark data by Artificial Analysis.

How Hy3 and Mistral Medium 3.5 score

🤝 Neck and neck on these criteria (4.5 vs 4.6 / 5).
CriterionHy3Mistral Medium 3.5
Cost-efficiency5.04.0
Context window4.04.0
Openness5.05.0
Self-hosting5.05.0
Multimodality3.55.0

Scores come from live data — output price (cost), context length, open vs closed weights (openness & self-hosting) and vision/tool support (multimodality). Raw task quality isn't scored here; it depends on your benchmark — see the verdict.

What each model is

Hy3 Open

Tencent · Open-weight

Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...

Mistral Medium 3.5 Open

Mistral AI · Open-weight

Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...

Other models in these families

These variants are tracked but not compared here — one page per family keeps the comparison readable.

Other variants tracked
Hy3 previewHunyuan A13B Instruct
Other variants tracked
Mistral Small 4Devstral 2 2512Mistral Large 3 2512Mistral Medium 3.1Mistral Small 3.1 24BMistral Medium 3Ministral 3 14B 2512Mistral Small 3.2 24BMinistral 3 8B 2512Mistral LargeMistral Large 2407Mistral Small 3SabaMinistral 3 3B 2512Mixtral 8x22B InstructCodestral 2508UncensoredVoxtral Small 24B 2507Mistral Nemo

Frequently asked questions

Hy3 vs Mistral Medium 3.5 — which is cheaper?

Hy3 is cheaper on output ($0.53/M vs $7.5/M).

Which has the larger context window?

Hy3 offers the larger context window (262K tokens).

Hy3 vs Mistral Medium 3.5 — which should I pick in 2026?

Choose Hy3 for the lower output price ($0.53/M vs $7.5/M). It also gives you the larger 262K context window.

People also compare

MiMo-V2.5-Pro vs Mistral Medium 3.5DeepSeek V4 Pro vs Mistral Medium 3.5Hy3 vs o1-proHy3 vs o3 ProHy3 vs GPT-4Hy3 vs o1

Explore more open-source AI

Browse the full open-source model leaderboard, thousands of tools and live benchmarks — all in one place.

Open the leaderboard →