The AI arena is free today

Open Superagent

Kimi K2-Thinking-0905 vs Seed 1.8

Kimi K2-Thinking-0905 and Seed 1.8 are closely matched at 36.0 and 34.2 on the LLM Stats Score. Seed 1.8 is 1.2x cheaper per token.

Moonshot AI · ByteDance · Updated for 2026

Which is better?

Kimi K2-Thinking-0905 and Seed 1.8 are closely matched on the overall LLM Stats Score at 36.0 and 34.2.

In the 7 individual benchmarks reported for both models, Seed 1.8 wins 5; this is a narrower head-to-head signal than the composite indexes.

On price, Seed 1.8 is roughly 1.2x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Kimi K2-Thinking-0905 also accepts a larger context window (262,144 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-Thinking-0905

  • you process long inputs — it offers a 262,144 token context window
  • you need open weights you can self-host or fine-tune

Choose Seed 1.8

  • you value its reported benchmark strengths — it wins 5 of 7 exact shared results
  • cost matters — it's about 1.2x cheaper per token
  • you want the most recent training data — it shipped Feb 2026

At a glance

The differences that matter most.

Core performance indexes
36.0
#79
34.2
#91
36.3
#73
33.5
#93
21.0
#90
19.8
#99
16.1
#78
14.9
#86
Cost, coverage & limits
Benchmark wins
2 of 7
5 of 7
Input price
$0.47 / M
$0.25 / M
Output price
$2.00 / M
$2.00 / M
Context window
262,144
256,000

Capability indexes

Additional strengths measured across groups of related public benchmarks

4 shared
Index
Kimi K2-Thinking-0905
Seed 1.8
38.0#27
32.5#57
16.2#37
19.3#23
19.3#86
32.7#13
18.5#80
32.7#13
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

21 reported for Kimi K2-Thinking-0905 · 57 for Seed 1.8

7 shared

Kimi K2-Thinking-0905 outperforms in 2 benchmarks (AIME 2025, LiveCodeBench v6), while Seed 1.8 is better at 5 benchmarks (BrowseComp, BrowseComp-zh, MMLU-Pro, Multi-SWE-Bench, SWE-Bench Verified).

Seed 1.8 shows notably better performance in the majority of benchmarks.

Tue Sep 08 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Seed 1.8 costs less

For input processing, Kimi K2-Thinking-0905 ($0.47/1M tokens) is 1.9x more expensive than Seed 1.8 ($0.25/1M tokens).

For output processing, Kimi K2-Thinking-0905 ($2.00/1M tokens) costs the same as Seed 1.8 ($2.00/1M tokens).

In conclusion, Kimi K2-Thinking-0905 is more expensive than Seed 1.8.*

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

Lowest available price from all providers
Tue Sep 08 2026 • llm-stats.com
Moonshot AI
Kimi K2-Thinking-0905
Input tokens$0.47
Output tokens$2.00
Best providerDeepinfra
ByteDance
Seed 1.8
Input tokens$0.25
Output tokens$2.00
Best providerDeepinfra
Notice missing or incorrect data?Start an Issue

Context Window

Maximum input and output token capacity

Kimi K2-Thinking-0905 accepts 262,144 input tokens compared to Seed 1.8's 256,000 tokens. Kimi K2-Thinking-0905 can generate longer responses up to 262,144 tokens, while Seed 1.8 is limited to 256,000 tokens.

Moonshot AI
Kimi K2-Thinking-0905
Input262,144 tokens
Output262,144 tokens
ByteDance
Seed 1.8
Input256,000 tokens
Output256,000 tokens
Tue Sep 08 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Seed 1.8 supports multimodal inputs, whereas Kimi K2-Thinking-0905 does not.

Seed 1.8 can handle both text and other forms of data like images, making it suitable for multimodal applications.

Kimi K2-Thinking-0905

Text
Images
Audio
Video

Seed 1.8

Text
Images
Audio
Video

License

Usage and distribution terms

Kimi K2-Thinking-0905 is licensed under MIT, while Seed 1.8 uses a proprietary license.

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

Kimi K2-Thinking-0905

MIT

Open weights

Seed 1.8

Proprietary

Closed source

Release Timeline

When each model was launched

Kimi K2-Thinking-0905 was released on 2025-09-05, while Seed 1.8 was released on 2026-02-17.

Seed 1.8 is 6 months newer than Kimi K2-Thinking-0905.

Kimi K2-Thinking-0905

Sep 5, 2025

1.0 years ago

Seed 1.8

Feb 17, 2026

6 months ago

5mo newer

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Provider Availability

Kimi K2-Thinking-0905 is available from DeepInfra, Novita, Fireworks. Seed 1.8 is available from DeepInfra.

Kimi K2-Thinking-0905

deepinfra logo
Deepinfra
Input Price:Input: $0.47/1MOutput Price:Output: $2.00/1M
novita logo
Novita
Input Price:Input: $0.48/1MOutput Price:Output: $2.00/1M
fireworks logo
Fireworks
Input Price:Input: $0.60/1MOutput Price:Output: $2.50/1M

Seed 1.8

deepinfra logo
Deepinfra
Input Price:Input: $0.25/1MOutput Price:Output: $2.00/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against Kimi K2-Thinking-0905 and Seed 1.8 side-by-side, then vote on the output you prefer.

Kimi K2-Thinking-0905
✓ Preferred
Seed 1.8
Open in Playground

FAQ

Common questions about Kimi K2-Thinking-0905 vs Seed 1.8.

Which is better, Kimi K2-Thinking-0905 or Seed 1.8?

Kimi K2-Thinking-0905 and Seed 1.8 are closely matched on the LLM Stats Score at 36.0 and 34.2. Kimi K2-Thinking-0905 is made by Moonshot AI and Seed 1.8 is made by ByteDance. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Kimi K2-Thinking-0905 compare to Seed 1.8 in benchmarks?

Kimi K2-Thinking-0905 scores AIME 2025: 100.0%, HMMT 2025: 97.5%, MMLU-Redux: 94.4%, FRAMES: 87.0%, MMLU-Pro: 84.6%. Seed 1.8 scores CountBench: 96.3%, AIME 2025: 94.3%, VLMsAreBlind: 93.0%, MMLU: 92.3%, AI2D: 89.1%.

Is Kimi K2-Thinking-0905 cheaper than Seed 1.8?

Seed 1.8 is 1.9x cheaper for input tokens. Kimi K2-Thinking-0905 costs $0.47/M input and $2.00/M output via deepinfra. Seed 1.8 costs $0.25/M input and $2.00/M output via deepinfra.

What are the context window sizes for Kimi K2-Thinking-0905 and Seed 1.8?

Kimi K2-Thinking-0905 supports 262K tokens and Seed 1.8 supports 256K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Kimi K2-Thinking-0905 and Seed 1.8?

Key differences include LLM Stats Score (36.0 vs 34.2), context window (262K vs 256K), input pricing ($0.47 vs $0.25/M), multimodal support (no vs yes), licensing (MIT vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes Kimi K2-Thinking-0905 and Seed 1.8?

Kimi K2-Thinking-0905 is developed by Moonshot AI and Seed 1.8 is developed by ByteDance.