Apertus 70B vs o4-mini-deep-research vs Kimi K2 Instruct 0905
Too close to call on our weighted score (Kimi K2 Instruct 0905 38, o4-mini-deep-research 37, Apertus 70B 32). The right pick depends on what you value most.
Swiss AI
Apertus 70B
32/100- ECI—
- Price$0.82 / $2.42
- Context66K
OpenAIo4-mini-deep-research
37/100- ECI—
- Price$1.80 / $7.20
- Context200K
Moonshot AIKimi K2 Instruct 0905
38/100- ECI—
- Price$0.60 / $2.50
- Context262K
Too close to call
It is close. Our weighted score puts them within 1 points (Kimi K2 Instruct 0905 38/100, o4-mini-deep-research 37/100, Apertus 70B 32/100), so choose by what matters most for your work: Kimi K2 Instruct 0905 on price and Kimi K2 Instruct 0905 for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceKimi K2 Instruct 0905Kimi K2 Instruct 0905 $1.07 · Apertus 70B $1.22 · o4-mini-deep-research $3.15 per 1M tokens (3:1 blend)
- Longest contextKimi K2 Instruct 0905Kimi K2 Instruct 0905 262,144 · o4-mini-deep-research 200,000 · Apertus 70B 65,536 tokens
- Widest inputso4-mini-deep-researchApertus 70B: Text · o4-mini-deep-research: Text, Images · Kimi K2 Instruct 0905: Text
- Self-hostingApertus 70B and Kimi K2 Instruct 0905Publishes downloadable weights (Apache-2.0 and Modified MIT)
| Measure | Weight | Apertus 70B | o4-mini-deep-research | Kimi K2 Instruct 0905 |
|---|---|---|---|---|
| Price | 50% | 44 | 25 | 46 |
| Inputs & features | 30% | 25 | 60 | 25 |
| Context window | 20% | 12 | 32 | 37 |
| Overall | 100% | 32/100 | 37/100 | 38/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Apertus 70BSwiss AI | o4-mini-deep-researchOpenAI | Kimi K2 Instruct 0905Moonshot AI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.82 | $1.80 | $0.60 (best) |
| Output | $2.42 (best) | $7.20 | $2.50 |
| Cached input | — | — | — |
| Blended (3:1) | $1.22 | $3.15 | $1.07 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 3 providers | Median of 1 providers | Median of 12 providers |
| Limits | |||
| Context window | 65,536 tokens | 200,000 tokens | 262,144 tokens (best) |
| Max output | 8,192 tokens | 100,000 tokens (best) | 98,304 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | Yes | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | OpenApache-2.0 | Proprietary | OpenModified MIT |
| API model ID | — | — | — |
| API providers | 3 | 1 | 13 (best) |
| Released | Sep 2, 2025 | Jun 26, 2025 | Sep 5, 2025 |
| Knowledge cutoff | Sep 2025 | May 2024 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Apertus 70B$13.04
- o4-mini-deep-research$32.40
- Kimi K2 Instruct 0905$11.00
Which should you choose?
Which is better: Apertus 70B, o4-mini-deep-research or Kimi K2 Instruct 0905?
It is close. Our weighted score puts them within 1 points (Kimi K2 Instruct 0905 38/100, o4-mini-deep-research 37/100, Apertus 70B 32/100), so choose by what matters most for your work: Kimi K2 Instruct 0905 on price and Kimi K2 Instruct 0905 for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Apertus 70B, o4-mini-deep-research or Kimi K2 Instruct 0905?
Kimi K2 Instruct 0905 is cheaper at $0.60 input / $2.50 output per million tokens (median across 12 API providers). Apertus 70B costs $0.82 input / $2.42 output per million tokens (median across 3 API providers); o4-mini-deep-research costs $1.80 input / $7.20 output per million tokens (median across 1 API provider). At a typical mix of three input tokens to one output token, that is $1.07 per million tokens for Kimi K2 Instruct 0905 versus $1.22 for Apertus 70B (1.1× as much) and $3.15 for o4-mini-deep-research (2.9× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Apertus 70B has not been scored yet, o4-mini-deep-research has not been scored yet and Kimi K2 Instruct 0905 has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Apertus 70B, o4-mini-deep-research and Kimi K2 Instruct 0905 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Kimi K2 Instruct 0905 has the largest context window at 262,144 tokens, against 200,000 for o4-mini-deep-research and 65,536 for Apertus 70B. Maximum output per response: Apertus 70B up to 8,192, o4-mini-deep-research up to 100,000, Kimi K2 Instruct 0905 up to 98,304 tokens.
Which can read images, PDFs, audio or video?
Apertus 70B accepts text; o4-mini-deep-research accepts text and images; Kimi K2 Instruct 0905 accepts text. o4-mini-deep-research handles the widest range of inputs.
Are any of these open source?
Apertus 70B and Kimi K2 Instruct 0905 publishes its weights (Apache-2.0 and Modified MIT) and can be self-hosted; o4-mini-deep-research is proprietary.
Which is newer?
Kimi K2 Instruct 0905 is the newest, released Sep 5, 2025. Apertus 70B came out Sep 2, 2025; o4-mini-deep-research came out Jun 26, 2025. Knowledge cutoff: Apertus 70B Sep 2025, o4-mini-deep-research May 2024.
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.
