| Signal | Kimi K2 0905 | Delta | Kimi K2 Thinking |
|---|---|---|---|
Capabilities | 50 | -17 | |
Benchmarks | 51 | -- | |
Pricing | 98 | -- | |
Context window size | 81 | -- | |
Recency | 94 | -6 | |
Output Capacity | 20 | -- | |
| Overall Result | 0 wins | of 6 | 2 wins |
Score History
52.2
current score
Kimi K2 Thinking
right now
53
current score
Moonshot AI
Moonshot AI
Kimi K2 0905 saves you $7.00/month
That's $84.00/year compared to Kimi K2 Thinking at your current usage level of 100K calls/month.
| Metric | Kimi K2 0905 | Kimi K2 Thinking | Winner |
|---|---|---|---|
| Overall Score | 52 | 53 | Kimi K2 Thinking |
| Rank | #149 | #148 | Kimi K2 Thinking |
| Quality Rank | #149 | #148 | Kimi K2 Thinking |
| Adoption Rank | #149 | #148 | Kimi K2 Thinking |
| Parameters | -- | -- | -- |
| Context Window | 131K | 131K | -- |
| Pricing | $0.40/$2.00/M | $0.47/$2.00/M | -- |
| Signal Scores | |||
| Capabilities | 50 | 67 | Kimi K2 Thinking |
| Benchmarks | 51 | 51 | Kimi K2 0905 |
| Pricing | 98 | 98 | Kimi K2 0905 |
| Context window size | 81 | 81 | Kimi K2 0905 |
| Recency | 94 | 100 | Kimi K2 Thinking |
| Output Capacity | 20 | 20 | Kimi K2 0905 |
Our score (0-100) is driven by benchmark performance (90%) from Arena Elo ratings, MMLU, GPQA, HumanEval, SWE-bench, and 15+ standardized evaluations. Capabilities and context window serve as tiebreakers (10%). Learn more about our methodology.
Scores 52/100 (rank #149), placing it in the top 49% of all 290 models tracked.
Scores 53/100 (rank #148), placing it in the top 49% of all 290 models tracked.
With only a 1-point gap, these models are in the same performance tier. The practical difference in output quality is minimal - your choice should depend on pricing, latency requirements, and specific feature needs.
Both models are priced similarly, so the decision comes down to quality and features rather than cost.
Both models have comparable response speeds. For most applications, the latency difference is negligible.
When latency matters most: Interactive chatbots, IDE code completion, real-time translation, and user-facing applications where response time directly impacts experience. For batch processing, background summarization, or offline analysis, latency is less critical.
Code generation & review
Higher benchmark score (0/100) indicates stronger performance on coding tasks like generating functions, debugging, and refactoring
Customer support chatbot
Faster response time (speed score 0/100) is critical for user-facing chat. Kimi K2 0905 also offers lower per-token costs for high-volume support
Long document analysis
Larger context window (131K tokens) can process longer documents, contracts, and research papers in a single pass
Batch data extraction
Lower output pricing ($2.00/M) reduces costs when processing thousands of records daily
Creative writing & content
Higher overall composite score (53/100) correlates with better nuance, coherence, and style in long-form content
Kimi K2 0905 and Kimi K2 Thinking are extremely close in overall performance (only 0.7999999999999972 points apart). Your best choice depends entirely on which specific strengths matter most for your use case.
Best for Quality
Kimi K2 0905
Marginally better benchmark scores; both are excellent
Best for Cost
Kimi K2 0905
3% lower pricing; better value at scale
Best for Reliability
Kimi K2 0905
Higher uptime and faster response speeds
Best for Prototyping
Kimi K2 0905
Stronger community support and better developer experience
Best for Production
Kimi K2 0905
Wider enterprise adoption and proven at scale
by Moonshot AI
| Capability | Kimi K2 0905 | Kimi K2 Thinking |
|---|---|---|
| Vision (Image Input) | ||
| Function Calling | ||
| Streaming | ||
| JSON Mode | ||
| Reasoningdiffers | ||
| Web Search | ||
| Image Output |
Moonshot AI
Moonshot AI
Kimi K2 0905 saves you $0.1260/month
That's 4% cheaper than Kimi K2 Thinking at 1,000 tokens/request and 100 requests/day.
Assumes 60% input / 40% output token ratio per request. Actual costs may vary based on your usage pattern.
| Parameter | Kimi K2 0905 | Kimi K2 Thinking |
|---|---|---|
| Context Window | 131K | 131K |
| Max Output Tokens | -- | -- |
| Open Source | Yes | Yes |
| Created | Sep 4, 2025 | Nov 6, 2025 |
Kimi K2 Thinking scores 53/100 (rank #148) compared to Kimi K2 0905's 52/100 (rank #149), giving it a 1-point advantage. Kimi K2 Thinking is the stronger overall choice, though Kimi K2 0905 may excel in specific areas like certain benchmarks.
Kimi K2 0905 is ranked #149 and Kimi K2 Thinking is ranked #148 out of 290+ AI models. Rankings use a composite score combining benchmark performance (90%) from MMLU, GPQA, HumanEval, SWE-bench, and 15+ standardized evaluations, with capabilities and context window as tiebreakers (10%). Scores update hourly.
Kimi K2 0905 is cheaper at $2.00/M output tokens vs Kimi K2 Thinking's $2.00/M output tokens - 1.0x more expensive. Input token pricing: Kimi K2 0905 at $0.40/M vs Kimi K2 Thinking at $0.47/M.
Kimi K2 0905 has a larger context window of 131,072 tokens compared to Kimi K2 Thinking's 131,072 tokens. A larger context window means the model can process longer documents and conversations.