Gemini 3 Pro Preview vs GPT-5.2 vs Kimi K2 Instruct
Gemini 3 Pro Preview comes out ahead, 65 to 60 and 53 on our weighted score, though Kimi K2 Instruct is 4.5× cheaper per token.
Price in rating: Included · Exclude price
- Our pick
Google
Gemini 3 Pro Preview
65/100- ECI153.0
- Price$2.00 / $12.00
- Context1.05M
OpenAI
GPT-5.2
60/100- ECI154.5
- Price$1.75 / $14.00
- Context400K
Moonshot AI
Kimi K2 Instruct
53/100- ECI141.3
- Price$0.57 / $2.30
- Context131K
Gemini 3 Pro Preview is our pick
Gemini 3 Pro Preview is the better all-round choice, scoring 65/100 against GPT-5.2 (60) and Kimi K2 Instruct (53). It leads on inputs & features and context window. Kimi K2 Instruct wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityGPT-5.2Capabilities Index (ECI): GPT-5.2 154.5 · Gemini 3 Pro Preview 153.0 · Kimi K2 Instruct 141.3
- Lowest priceKimi K2 InstructKimi K2 Instruct $1.00 · Gemini 3 Pro Preview $4.50 · GPT-5.2 $4.81 per 1M tokens (3:1 blend)
- Longest contextGemini 3 Pro PreviewGemini 3 Pro Preview 1,048,576 · GPT-5.2 400,000 · Kimi K2 Instruct 131,072 tokens
- Widest inputsGemini 3 Pro PreviewGemini 3 Pro Preview: Text, Images, PDFs, Audio, Video · GPT-5.2: Text, Images · Kimi K2 Instruct: Text
- Self-hostingKimi K2 InstructPublishes downloadable weights (Modified MIT)
| Measure | Weight | Gemini 3 Pro Preview | GPT-5.2 | Kimi K2 Instruct |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 80 | 82 | 69 |
| Price | 25% | 18 | 17 | 48 |
| Inputs & features | 15% | 90 | 70 | 25 |
| Context window | 10% | 65 | 50 | 32 |
| Overall | 100% | 65/100 | 60/100 | 53/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 153.0 | 154.5 (best) | 141.3 |
| ECI rank | #55 of 158 | #51 of 158 (best) | #108 of 158 |
| GPQA DiamondGraduate-level science questions | 92.6% (best) | 91.4% | — |
| FrontierMath Tiers 1–3Research-level mathematics | — | 67.4% | — |
| OTIS Mock AIME 2024–2025Competition mathematics | 91.4% | 96.1% (best) | — |
| SWE-bench VerifiedFixing real GitHub issues | 72.9% | 73.8% (best) | — |
| SimpleQA VerifiedShort factual questions | — | 37.1% | — |
| Price per million tokens | |||
| Input | $2.00 | $1.75 | $0.57 (best) |
| Output | $12.00 | $14.00 | $2.30 (best) |
| Cached input | — | $0.175 | — |
| Blended (3:1) | $4.50 | $4.81 | $1.00 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 6 providers | Official OpenAI API | Median of 11 providers |
| Limits | |||
| Context window | 1,048,576 tokens (best) | 400,000 tokens | 131,072 tokens |
| Max output | 65,536 tokens | 128,000 tokens (best) | 32,768 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | No |
| PDFs | Yes | No | No |
| Audio | Yes | No | No |
| Video | Yes | No | No |
| Reasoning | Yes | Yeslow · medium · high · xhigh | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | Yes | No |
| Availability | |||
| Weights | Proprietary | Proprietary | OpenModified MIT |
| API model ID | — | gpt-5.2 | — |
| API providers | 8 | 36 (best) | 11 |
| Released | Nov 18, 2025 | Dec 11, 2025 | Jul 11, 2025 |
| Knowledge cutoff | Jan 2025 | Aug 31, 2025 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
Gemini 3 Pro Preview$44.00
GPT-5.2$45.50
Kimi K2 Instruct$10.30
Which should you choose?
Which is better: Gemini 3 Pro Preview, GPT-5.2 or Kimi K2 Instruct?
Gemini 3 Pro Preview is the better all-round choice, scoring 65/100 against GPT-5.2 (60) and Kimi K2 Instruct (53). It leads on inputs & features and context window. Kimi K2 Instruct wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, Gemini 3 Pro Preview, GPT-5.2 or Kimi K2 Instruct?
Kimi K2 Instruct is cheaper at $0.57 input / $2.30 output per million tokens (median across 11 API providers). Gemini 3 Pro Preview costs $2.00 input / $12.00 output per million tokens (median across 6 API providers); GPT-5.2 costs $1.75 input / $14.00 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $1.00 per million tokens for Kimi K2 Instruct versus $4.50 for Gemini 3 Pro Preview (4.5× as much) and $4.81 for GPT-5.2 (4.8× as much).
Which scores higher on benchmarks?
GPT-5.2 scores higher on the Capabilities Index (ECI): GPT-5.2 154.5 (#51 of 158), Gemini 3 Pro Preview 153.0 (#55 of 158) and Kimi K2 Instruct 141.3 (#108 of 158). The confidence ranges of the top two overlap (152.9–156.3 vs 151.5–154.8), so treat the gap as small.
Which is better for coding?
There are no published SWE-bench Verified results for Kimi K2 Instruct yet, so there is no like-for-like coding score. On overall capability, GPT-5.2 leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.
Which has the bigger context window?
Gemini 3 Pro Preview has the largest context window at 1,048,576 tokens, against 400,000 for GPT-5.2 and 131,072 for Kimi K2 Instruct. Maximum output per response: Gemini 3 Pro Preview up to 65,536, GPT-5.2 up to 128,000, Kimi K2 Instruct up to 32,768 tokens.
Which can read images, PDFs, audio or video?
Gemini 3 Pro Preview accepts text, images, PDFs, audio and video; GPT-5.2 accepts text and images; Kimi K2 Instruct accepts text. Gemini 3 Pro Preview handles the widest range of inputs.
Are any of these open source?
Kimi K2 Instruct publishes its weights (Modified MIT) and can be self-hosted; Gemini 3 Pro Preview and GPT-5.2 is proprietary.
Which is newer?
GPT-5.2 is the newest, released Dec 11, 2025. Gemini 3 Pro Preview came out Nov 18, 2025; Kimi K2 Instruct came out Jul 11, 2025. Knowledge cutoff: Gemini 3 Pro Preview Jan 2025, GPT-5.2 Aug 31, 2025.
How do you decide the winner?
This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 18, 2026.