Moonshot AIKimi K2 0905
44 cheaper ranked models score higher: DeepSeek V4.1 Flash 1017 (59.2, $0.525), DeepSeek V4.1 Flash (59.0, $0.518), GPT-5.6 Luna (58.5, $0.450) and 41 more.
As of 6 Nov 2026, Kimi K2 0905 by Moonshot AI scores 51.8 on the BenchLeader Index (587 ranked configurations), with its best results in human preference (60) and weakest in instruction following (44). API pricing: $0.600 per 1M input tokens and $2.50 per 1M output tokens ($1.07 blended at 3:1), list price on 6 Nov 2026. It streams about 42 tokens/s (Artificial Analysis). Released 5 Sep 2025.
Key facts (as of 6 Nov 2026)
- Kimi K2 0905 API pricing: $0.600 per 1M input tokens and $2.50 per 1M output tokens ($1.07 blended at 3:1) (list price, 6 Nov 2026).
- Kimi K2 0905 streams about 42 tokens/s (Artificial Analysis, 6 Nov 2026).
- Kimi K2 0905 was released on 5 Sep 2025 (Moonshot AI).
In detail. Kimi K2 0905 is a Moonshot AI open-weights model, released 5 Sep 2025. Its best configuration ranks #292 of 587 on the BenchLeader Index at 51.8 ±4.8, in the upper half. The ± is the point: 273 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in human preference (60) and lowest in instruction following (44). At $1.07 per million tokens blended it is mid-priced. Output speed of 42 tokens per second puts it in the slowest quarter, with a first answer in 2.6 s. Last measured 28 Oct 2026.
- Blended price
- $1.07/M
- $0.600 in · $2.50 out
- via Artificial Analysis
- Output speed
- 42 tok/s
- measured by Artificial Analysis
- First answer
- 2.59 s
- first token 0.71 s
- Context
- 262k
- via models.dev
- Full answer
- 92 s
- median, reasoning included
- Cost per task
- –
- Tokens per task
- –
- Released
- 5 Sep 2025
- Overall index52
- Reasoning54
- Coding50
- Agents & tools54
- Maths59
- Instruction following44
- Human preference60
- Long context45
Versions
Moonshot AI has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Kimi K2 0905this page | 5 Sep 2025 | 51.8 | #292 |
| Kimi K2 | 11 Jul 2025 | 51.0 | #317 |
How its index has moved
10 Sep 2026 to 28 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1438#129 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 76.7%#234 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 6.4%#406 | Artificial Analysis | |
| Kagi LLM Benchmark | 52.2%#76 | Kagi LLM Benchmark | |
| CritPtnot in index | 0.0%#328 | Artificial Analysis | |
| ForecastBenchnot in index | 62.7%#52 | ForecastBench |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| WeirdML | 36.7%#122 | WeirdML | |
| Arena Coding | 1468#122 | Arena (LMArena) | |
| SWE-bench Verified (any scaffold)not in index | 71.2%#13 | SWE-bench | |
| ALE-Benchnot in index | 267.1#124 | Epoch AI Benchmarking Hub | |
| LiveCodeBench (AA)not in index | 61.0%#108 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 23.5%#152 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 73.4%#146 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Maths | 1416#129 | Arena (LMArena) |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 41.7%#247 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1390#152 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1419#127 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1382#135 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1404#151 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1406#154 | Arena (LMArena) |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Fiction.LiveBench 120k | 40.6%#25 | Fiction.live | |
| AA-LCR v1.1 | 53.0%#321 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2not in index | 15.3#294 | Artificial Analysis | |
| AA Progress Index v1not in index | 220.8#347 | Artificial Analysis |
API pricing: where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| NovitaAI | 33 tok/s | 0.81 s | $0.600 | $2.50 | 262k | fp8 |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0010 | – | 9.8 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0087 | – | 17.0 s |
| Code edit | 6,000 / 1,500 | $0.0074 | – | 38.6 s |
| Agentic coding session | 60,000 / 4,000 | $0.046 | – | 1.6 min |
| Structured extraction | 2,000 / 200 | $0.0017 | – | 7.4 s |
More from Moonshot AI
See also
Embed or cite
Show where Kimi K2 0905 ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.
[](https://www.benchleader.com/models/kimi-k2-0905)Cite this
The ranking changes daily, so a reference carries the data date. APA style:
BenchLeader. (2026, November 6). Kimi K2 0905 on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 6, 2026, from https://www.benchleader.com/models/kimi-k2-0905
BibTeX:
@misc{benchleader_kimi_k2_0905_20261106,
author = {{BenchLeader}},
title = {{Kimi K2 0905 on the BenchLeader Index (method v1.4)}},
year = {2026},
month = nov,
howpublished = {\url{https://www.benchleader.com/models/kimi-k2-0905}},
note = {Data as of 2026-11-06}
}Every badge style, the embeddable top-10 widget and how to cite the Index
Questions people ask
- What are Kimi K2 0905's benchmark scores?
- As of 6 Nov 2026, Kimi K2 0905 scores 51.8 ±4.8 on the BenchLeader Index, strongest in human preference (60) and weakest in instruction following (44). That rests on independent benchmark results only, and every score on this page links to its publisher.
- How does Kimi K2 0905 rank?
- Kimi K2 0905 is #292 of 587 ranked configurations on 6 Nov 2026. 273 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 69.3.
- How much does Kimi K2 0905 cost per 1M tokens?
- As of 6 Nov 2026, Kimi K2 0905 costs $0.600 per 1M input tokens and $2.50 per 1M output tokens, $1.07 blended (three input tokens per output token).
- When was Kimi K2 0905 released?
- Moonshot AI released Kimi K2 0905 on 5 Sep 2025.
Data as of 6 Nov 2026. Compare with another model.
Cite as: BenchLeader, “Kimi K2 0905: benchmarks, pricing, speed and rank”, data as of 6 Nov 2026, https://www.benchleader.com/models/kimi-k2-0905.