GPT-5.6 Terra vs Kimi K3 vs Muse Spark 1.2
Muse Spark 1.2 comes out ahead, 71 to 67 and 65 on our weighted score, and it is the cheaper option too.
OpenAIGPT-5.6 Terra
67/100- ECI160.1
- Price$2.00 / $12.00
- Context1.05M
Moonshot AIKimi K3
65/100- ECI158.3
- Price$3.00 / $15.00
- Context1.05M
- Our pick
MetaMuse Spark 1.2
71/100- ECI155.2
- Price$1.25 / $4.25
- Context1.05M
Muse Spark 1.2 is our pick
Muse Spark 1.2 is the better all-round choice, scoring 71/100 against GPT-5.6 Terra (67) and Kimi K3 (65). It leads on price and inputs & features. GPT-5.6 Terra wins on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityGPT-5.6 TerraCapabilities Index (ECI): GPT-5.6 Terra 160.1 · Kimi K3 158.3 · Muse Spark 1.2 155.2
- Lowest priceMuse Spark 1.2Muse Spark 1.2 $2.00 · GPT-5.6 Terra $4.50 · Kimi K3 $6.00 per 1M tokens (3:1 blend)
- Longest contextAbout the sameGPT-5.6 Terra 1,050,000 · Kimi K3 1,048,576 · Muse Spark 1.2 1,048,576 tokens
- Widest inputsMuse Spark 1.2GPT-5.6 Terra: Text, Images, PDFs · Kimi K3: Text, Images, Video · Muse Spark 1.2: Text, Images, PDFs, Audio, Video
- Self-hostingKimi K3Publishes downloadable weights
| Measure | Weight | GPT-5.6 Terra | Kimi K3 | Muse Spark 1.2 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 89 | 87 | 83 |
| Price | 25% | 18 | 12 | 34 |
| Inputs & features | 15% | 80 | 80 | 100 |
| Context window | 10% | 61 | 61 | 61 |
| Overall | 100% | 67/100 | 65/100 | 71/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | GPT-5.6 TerraOpenAI | Kimi K3Moonshot AI | Muse Spark 1.2Meta |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 160.1 (best) | 158.3 | 155.2 |
| ECI rank | #15 of 165 (best) | #20 of 165 | #38 of 165 |
| GPQA DiamondGraduate-level science questions | 93.3% (best) | 93.1% | — |
| FrontierMath Tiers 1–3Research-level mathematics | 86.0% (best) | 72.2% | — |
| OTIS Mock AIME 2024–2025Competition mathematics | 99.7% (best) | 97.2% | — |
| SimpleQA VerifiedShort factual questions | 43.2% | 50.6% | 60.3% (best) |
| Price per million tokens | |||
| Input | $2.00 | $3.00 | $1.25 (best) |
| Output | $12.00 | $15.00 | $4.25 (best) |
| Cached input | $0.20 | $0.30 | $0.15 (best) |
| Blended (3:1) | $4.50 | $6.00 | $2.00 (best) |
| Long-context rate | Over 272K: $4.00 / $18.00 | Same rate | Same rate |
| Price source | Official OpenAI API | Official Moonshot AI API | Official Meta API |
| Limits | |||
| Context window | 1,050,000 tokens (best) | 1,048,576 tokens | 1,048,576 tokens |
| Max output | 128,000 tokens | 131,072 tokens (best) | 131,072 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | Yes | No | Yes |
| Audio | No | No | Yes |
| Video | No | Yes | Yes |
| Reasoning | Yeslow · medium · high · xhigh · max | Yeslow · high · max | Yesminimal · low · medium · high · xhigh |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Yes |
| Availability | |||
| Weights | Proprietary | Open | Proprietary |
| API model ID | gpt-5.6-terra | kimi-k3 | muse-spark-1.2 |
| API providers | 45 | 76 (best) | 18 |
| Released | Jul 9, 2026 | Jul 16, 2026 | Aug 5, 2026 |
| Knowledge cutoff | Feb 16, 2026 | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- GPT-5.6 Terra$44.00
- Kimi K3$60.00
- Muse Spark 1.2$21.00
Which should you choose?
Which is better: GPT-5.6 Terra, Kimi K3 or Muse Spark 1.2?
Muse Spark 1.2 is the better all-round choice, scoring 71/100 against GPT-5.6 Terra (67) and Kimi K3 (65). It leads on price and inputs & features. GPT-5.6 Terra wins on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, GPT-5.6 Terra, Kimi K3 or Muse Spark 1.2?
Muse Spark 1.2 is cheaper at $1.25 input / $4.25 output per million tokens (official Meta API price). GPT-5.6 Terra costs $2.00 input / $12.00 output per million tokens (official OpenAI API price); Kimi K3 costs $3.00 input / $15.00 output per million tokens (official Moonshot AI API price). At a typical mix of three input tokens to one output token, that is $2.00 per million tokens for Muse Spark 1.2 versus $4.50 for GPT-5.6 Terra (2.3× as much) and $6.00 for Kimi K3 (3× as much).
Which scores higher on benchmarks?
GPT-5.6 Terra scores higher on the Capabilities Index (ECI): GPT-5.6 Terra 160.1 (#15 of 165), Kimi K3 158.3 (#20 of 165) and Muse Spark 1.2 155.2 (#38 of 165). The confidence ranges of the top two overlap (157.8–163.4 vs 155.8–161.4), so treat the gap as small. On individual benchmarks: SimpleQA Verified — Muse Spark 1.2 60.3%, Kimi K3 50.6%, GPT-5.6 Terra 43.2%.
Which is better for coding?
There are no published SWE-bench Verified results for GPT-5.6 Terra, Kimi K3 and Muse Spark 1.2 yet, so there is no like-for-like coding score. On overall capability, GPT-5.6 Terra 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?
Their context windows are effectively the same size: GPT-5.6 Terra 1,050,000, Kimi K3 1,048,576 and Muse Spark 1.2 1,048,576 tokens. Maximum output per response: GPT-5.6 Terra up to 128,000, Kimi K3 up to 131,072, Muse Spark 1.2 up to 131,072 tokens.
Which can read images, PDFs, audio or video?
GPT-5.6 Terra accepts text, images and PDFs; Kimi K3 accepts text, images and video; Muse Spark 1.2 accepts text, images, PDFs, audio and video. Muse Spark 1.2 handles the widest range of inputs.
Are any of these open source?
Kimi K3 publishes its weights and can be self-hosted; GPT-5.6 Terra and Muse Spark 1.2 is proprietary.
Which is newer?
Muse Spark 1.2 is the newest, released Aug 5, 2026. Kimi K3 came out Jul 16, 2026; GPT-5.6 Terra came out Jul 9, 2026. Knowledge cutoff: GPT-5.6 Terra Feb 16, 2026.
How do you decide the winner?
Each model gets a 0–100 score on 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 2, 2026.
