Kimi K2 0905 vs Kimi K2 Thinking
Prices checked
Kimi K2 0905 takes 1 of 5 rounds against Kimi K2 Thinking. It is also the cheaper of the two to run, at $0.75 per million tokens against $0.86.
Moonshot AIWins 1 roundKimi K2 0905
262k context · tool calling · open weights
$0.75/M tokensCheapest price, at NanoGPT · $0.35 per 1,000 pages
Moonshot AIKimi K2 Thinking
262k context · tool calling · open weights
$0.86/M tokensCheapest price, at Helicone · $0.41 per 1,000 pages
Maker's price in (lower wins)
$0.60/M
$0.60/M
Maker's price out (lower wins)
$2.50/M
$2.50/M
Cheapest price (lower wins)
$0.75/M at NanoGPT
$0.86/M at Helicone
Maker's price, blended (lower wins)
$1.08/M
$1.08/M
Cost per 1,000 pages (lower wins)
$0.35
$0.41
Context
262k tokens
262k tokens
Open weights
yes
yes
Tool calling
yes
yes
Structured output
yes
yes
Reads images
no
no
Reasoning
no
yes
Released
4 Sept 2025
6 Nov 2025
Providers selling it
26
21
Price we measure against
OpenRouter
OpenRouter
Trains on your prompts (no is better)
not published
not published
Providers selling both · 15
Helicone$0.88 / $0.86
NanoGPT$0.75 / $1.08
OpenCode$0.93 / $0.93
TrustedRouter$1.13 / $0.79
Qiniu$1.01 / $1.01
io.net Intelligence$1.00 / $1.08
302.AI$1.10 / $1.01
Baseten$1.08 / $1.08
Kimi K2 0905, then Kimi K2 Thinking, per million tokens; the 8 cheapest of 15.
Go further
Add a third or fourth model, or change the weighting, in the tool.
What moved this week
AI/ML API · 49% cheaper ·
$0.60 / $2.43 $1.38 / $4.13
Frequently asked
- Which is cheaper, Kimi K2 0905 or Kimi K2 Thinking?
- Kimi K2 0905, at $0.75 per million tokens against $0.86 for Kimi K2 Thinking, blended 3 in : 1 out, read on 31 Oct 2026.
- Where is each one cheapest?
- Kimi K2 0905: $0.75 at NanoGPT. Kimi K2 Thinking: $0.86 at Helicone.
- Can I buy both from one provider?
- Yes. 15 providers sell both: Helicone, NanoGPT, OpenCode, TrustedRouter, Qiniu, io.net Intelligence, and more. The cheapest of them for the two together is Helicone.
- Which holds more context?
- Both hold 262k tokens in one call.
Prices exclude tax, read 31 Oct 2026 from each provider's own catalog. They are per M tokens, blended 3 in : 1 out, and a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.
Related comparisons