Kimi K2.5 vs Mercury 3
Kimi K2.5 and Mercury 3 are closely matched at 38.5 and 36.1 on the LLM Stats Score. Mercury 3 is 7.4x cheaper per token.
Moonshot AI · Inception · Updated for 2026
Which is better?
Kimi K2.5 and Mercury 3 are closely matched on the overall LLM Stats Score at 38.5 and 36.1.
In the 3 individual benchmarks reported for both models, Mercury 3 wins 2; this is a narrower head-to-head signal than the composite indexes.
On price, Mercury 3 is roughly 7.4x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.
Mercury 3 also accepts a larger context window (1,000,000 input tokens), making it the stronger choice for long documents and large codebases.
Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.
Choose Kimi K2.5
- you need open weights you can self-host or fine-tune
Choose Mercury 3
- you value its reported benchmark strengths — it wins 2 of 3 exact shared results
- cost matters — it's about 7.4x cheaper per token
- you process long inputs — it offers a 1,000,000 token context window
- you want the most recent training data — it shipped Oct 2026
At a glance
The differences that matter most.
Capability indexes
Additional strengths measured across groups of related public benchmarks
Individual benchmarks
40 reported for Kimi K2.5 · 12 for Mercury 3
Kimi K2.5 outperforms in 1 benchmarks (DeepSearchQA), while Mercury 3 is better at 2 benchmarks (AA-LCR, GPQA).
Mercury 3 shows notably better performance in the majority of benchmarks.
Human preference
Blind head-to-head votes and playground preference scores
Pricing Analysis
Price comparison per million tokens
For input processing, Kimi K2.5 ($0.60/1M tokens) is 6.0x more expensive than Mercury 3 ($0.10/1M tokens).
For output processing, Kimi K2.5 ($3.00/1M tokens) is 8.6x more expensive than Mercury 3 ($0.35/1M tokens).
In conclusion, Kimi K2.5 is more expensive than Mercury 3.*
* Using a 3:1 ratio of input to output tokens
Context Window
Maximum input and output token capacity
Mercury 3 accepts 1,000,000 input tokens compared to Kimi K2.5's 262,100 tokens. Kimi K2.5 can generate longer responses up to 262,100 tokens, while Mercury 3 is limited to 131,072 tokens.
Input capabilities
Documented input modalities across available providers
Kimi K2.5 supports multimodal inputs, whereas Mercury 3 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
Mercury 3
License
Usage and distribution terms
Kimi K2.5 is licensed under MIT, while Mercury 3 uses a proprietary license.
License differences may affect how you can use these models in commercial or open-source projects.
MIT
Open weights
Proprietary
Closed source
Release Timeline
When each model was launched
Kimi K2.5 was released on 2026-01-27, while Mercury 3 was released on 2026-10-19.
Mercury 3 is 9 months newer than Kimi K2.5.
Jan 27, 2026
8 months ago
Oct 19, 2026
0 days ago
8mo newerKnowledge Cutoff
When training data ends
Neither model specifies a knowledge cutoff date.
Unable to compare the recency of their training data.
Provider Availability
Kimi K2.5 is available from Fireworks, Moonshot AI. Mercury 3 is available from Inception.
Kimi K2.5
Mercury 3
Outputs Comparison
Judge for yourself.
Run your own prompts against Kimi K2.5 and Mercury 3 side-by-side, then vote on the output you prefer.
FAQ
Common questions about Kimi K2.5 vs Mercury 3.