Model Comparison

Kimi K2.5 vs Nemotron 3 Nano (30B A3B)

Kimi K2.5 significantly outperforms across most benchmarks. Nemotron 3 Nano (30B A3B) is 10.2x cheaper per token.

Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

7 benchmarks

Kimi K2.5 outperforms in 6 benchmarks (GPQA, Humanity's Last Exam, LiveCodeBench v6, MMLU-Pro, SciCode, SWE-Bench Verified), while Nemotron 3 Nano (30B A3B) is better at 1 benchmark (AIME 2025).

Kimi K2.5 significantly outperforms across most benchmarks.

Tue Apr 07 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Nemotron 3 Nano (30B A3B) costs less

For input processing, Kimi K2.5 ($0.60/1M tokens) is 10.0x more expensive than Nemotron 3 Nano (30B A3B) ($0.06/1M tokens).

For output processing, Kimi K2.5 ($2.50/1M tokens) is 10.4x more expensive than Nemotron 3 Nano (30B A3B) ($0.24/1M tokens).

In conclusion, Kimi K2.5 is more expensive than Nemotron 3 Nano (30B A3B).*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Tue Apr 07 2026 • llm-stats.com
Moonshot AI
Kimi K2.5
Input tokens$0.60
Output tokens$2.50
Best providerFireworks
NVIDIA
Nemotron 3 Nano (30B A3B)
Input tokens$0.06
Output tokens$0.24
Best providerDeepinfra
Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

968.0B diff

Kimi K2.5 has 968.0B more parameters than Nemotron 3 Nano (30B A3B), making it 3025.0% larger.

Moonshot AI
Kimi K2.5
1000.0Bparameters
NVIDIA
Nemotron 3 Nano (30B A3B)
32.0Bparameters
1000.0B
Kimi K2.5
32.0B
Nemotron 3 Nano (30B A3B)

Context Window

Maximum input and output token capacity

Nemotron 3 Nano (30B A3B) accepts 262,144 input tokens compared to Kimi K2.5's 262,100 tokens. Nemotron 3 Nano (30B A3B) can generate longer responses up to 262,144 tokens, while Kimi K2.5 is limited to 262,100 tokens.

Moonshot AI
Kimi K2.5
Input262,100 tokens
Output262,100 tokens
NVIDIA
Nemotron 3 Nano (30B A3B)
Input262,144 tokens
Output262,144 tokens
Tue Apr 07 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Kimi K2.5 supports multimodal inputs, whereas Nemotron 3 Nano (30B A3B) does not.

Kimi K2.5 can handle both text and other forms of data like images, making it suitable for multimodal applications.

Kimi K2.5

Text
Images
Audio
Video

Nemotron 3 Nano (30B A3B)

Text
Images
Audio
Video

License

Usage and distribution terms

Kimi K2.5 is licensed under MIT, while Nemotron 3 Nano (30B A3B) uses NVIDIA Open Model License Agreement .

License differences may affect how you can use these models in commercial or open-source projects.

Kimi K2.5

MIT

Open weights

Nemotron 3 Nano (30B A3B)

NVIDIA Open Model License Agreement

Open weights

Release Timeline

When each model was launched

Kimi K2.5 was released on 2026-01-27, while Nemotron 3 Nano (30B A3B) was released on 2025-12-15.

Kimi K2.5 is 1 month newer than Nemotron 3 Nano (30B A3B).

Kimi K2.5

Jan 27, 2026

2 months ago

1mo newer
Nemotron 3 Nano (30B A3B)

Dec 15, 2025

3 months ago

Knowledge Cutoff

When training data ends

Nemotron 3 Nano (30B A3B) has a documented knowledge cutoff of 2025-11-28, while Kimi K2.5's cutoff date is not specified.

We can confirm Nemotron 3 Nano (30B A3B)'s training data extends to 2025-11-28, but cannot make a direct comparison without Kimi K2.5's cutoff date.

Kimi K2.5

—

Nemotron 3 Nano (30B A3B)

Nov 2025

Provider Availability

Kimi K2.5 is available from Fireworks. Nemotron 3 Nano (30B A3B) is available from DeepInfra.

Kimi K2.5

fireworks logo
Fireworks
Input Price:Input: $0.60/1MOutput Price:Output: $2.50/1M

Nemotron 3 Nano (30B A3B)

deepinfra logo
Deepinfra
Input Price:Input: $0.06/1MOutput Price:Output: $0.24/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

Moonshot AI
Supports multimodal inputs
Higher GPQA score (87.6% vs 75.0%)
Higher Humanity's Last Exam score (50.2% vs 15.5%)
Higher LiveCodeBench v6 score (85.0% vs 68.3%)
Higher MMLU-Pro score (87.1% vs 78.3%)
Higher SciCode score (48.7% vs 33.3%)
Higher SWE-Bench Verified score (76.8% vs 38.8%)
NVIDIA
Larger context window (262,144 tokens)
Less expensive input tokens
Less expensive output tokens
Higher AIME 2025 score (99.2% vs 96.1%)

Detailed Comparison

AI Model Comparison Table
Feature
Moonshot AI
Kimi K2.5
NVIDIA
Nemotron 3 Nano (30B A3B)

FAQ

Common questions about Kimi K2.5 vs Nemotron 3 Nano (30B A3B)

Kimi K2.5 significantly outperforms across most benchmarks. Kimi K2.5 is made by Moonshot AI and Nemotron 3 Nano (30B A3B) is made by NVIDIA. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.
Kimi K2.5 scores AIME 2025: 96.1%, HMMT 2025: 95.4%, InfoVQAtest: 92.6%, OCRBench: 92.3%, MathVista-Mini: 90.1%. Nemotron 3 Nano (30B A3B) scores AIME 2025: 99.2%, WMT24++: 86.2%, MMLU-Pro: 78.3%, GPQA: 75.0%, LiveCodeBench v6: 68.3%.
Nemotron 3 Nano (30B A3B) is 10.0x cheaper for input tokens. Kimi K2.5 costs $0.60/M input and $2.50/M output via fireworks. Nemotron 3 Nano (30B A3B) costs $0.06/M input and $0.24/M output via deepinfra.
Kimi K2.5 supports 262K tokens and Nemotron 3 Nano (30B A3B) supports 262K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.
Key differences include context window (262K vs 262K), input pricing ($0.60 vs $0.06/M), multimodal support (yes vs no), licensing (MIT vs NVIDIA Open Model License Agreement ). See the full comparison above for benchmark-by-benchmark results.
Kimi K2.5 is developed by Moonshot AI and Nemotron 3 Nano (30B A3B) is developed by NVIDIA.