DeepSeek-V3.1 vs Kimi K2 Instruct vs GPT OSS 120B
Too close to call on our weighted score (GPT OSS 120B 58, DeepSeek-V3.1 56, Kimi K2 Instruct 50). The right pick depends on what you value most.
DeepSeekDeepSeek-V3.1
56/100- ECI139.8
- Price$0.27 / $1.00
- Context131K
Moonshot AIKimi K2 Instruct
50/100- ECI139.9
- Price$0.57 / $2.30
- Context131K
OpenAIGPT OSS 120B
58/100- ECI139.8
- Price$0.15 / $0.60
- Context131K
Too close to call
It is close. Our weighted score puts them within 3 points (GPT OSS 120B 58/100, DeepSeek-V3.1 56/100, Kimi K2 Instruct 50/100), so choose by what matters most for your work: Kimi K2 Instruct for raw capability and GPT OSS 120B on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityKimi K2 InstructCapabilities Index (ECI): Kimi K2 Instruct 139.9 · DeepSeek-V3.1 139.8 · GPT OSS 120B 139.8
- Lowest priceGPT OSS 120BGPT OSS 120B $0.263 · DeepSeek-V3.1 $0.453 · Kimi K2 Instruct $1.00 per 1M tokens (3:1 blend)
- Longest contextAbout the sameDeepSeek-V3.1 131,072 · Kimi K2 Instruct 131,072 · GPT OSS 120B 131,072 tokens
- Widest inputsSame inputsDeepSeek-V3.1: Text · Kimi K2 Instruct: Text · GPT OSS 120B: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | DeepSeek-V3.1 | Kimi K2 Instruct | GPT OSS 120B |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 64 | 64 | 64 |
| Price | 25% | 63 | 48 | 74 |
| Inputs & features | 15% | 35 | 25 | 35 |
| Context window | 10% | 25 | 25 | 25 |
| Overall | 100% | 56/100 | 50/100 | 58/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | DeepSeek-V3.1DeepSeek | Kimi K2 InstructMoonshot AI | GPT OSS 120BOpenAI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 139.8 | 139.9 (best) | 139.8 |
| ECI rank | #113 of 165 | #112 of 165 (best) | #114 of 165 |
| GPQA DiamondGraduate-level science questions | — | — | 75.8% |
| OTIS Mock AIME 2024–2025Competition mathematics | — | — | 88.9% |
| Price per million tokens | |||
| Input | $0.27 | $0.57 | $0.15 (best) |
| Output | $1.00 | $2.30 | $0.60 (best) |
| Cached input | — | — | — |
| Blended (3:1) | $0.453 | $1.00 | $0.263 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 19 providers | Median of 11 providers | Median of 65 providers |
| Limits | |||
| Context window | 131,072 tokens | 131,072 tokens | 131,072 tokens |
| Max output | 8,192 tokens | 32,768 tokens (best) | 32,768 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | No | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | OpenMIT License | OpenModified MIT | Open |
| API model ID | — | — | — |
| API providers | 19 | 11 | 67 (best) |
| Released | Aug 21, 2025 | Jul 11, 2025 | Aug 5, 2025 |
| Knowledge cutoff | — | — | Jun 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- DeepSeek-V3.1$4.70
- Kimi K2 Instruct$10.30
- GPT OSS 120B$2.70
Which should you choose?
Which is better: DeepSeek-V3.1, Kimi K2 Instruct or GPT OSS 120B?
It is close. Our weighted score puts them within 3 points (GPT OSS 120B 58/100, DeepSeek-V3.1 56/100, Kimi K2 Instruct 50/100), so choose by what matters most for your work: Kimi K2 Instruct for raw capability and GPT OSS 120B on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, DeepSeek-V3.1, Kimi K2 Instruct or GPT OSS 120B?
GPT OSS 120B is cheaper at $0.15 input / $0.60 output per million tokens (median across 65 API providers). DeepSeek-V3.1 costs $0.27 input / $1.00 output per million tokens (median across 19 API providers); Kimi K2 Instruct costs $0.57 input / $2.30 output per million tokens (median across 11 API providers). At a typical mix of three input tokens to one output token, that is $0.263 per million tokens for GPT OSS 120B versus $0.453 for DeepSeek-V3.1 (1.7× as much) and $1.00 for Kimi K2 Instruct (3.8× as much).
Which scores higher on benchmarks?
Kimi K2 Instruct scores higher on the Capabilities Index (ECI): Kimi K2 Instruct 139.9 (#112 of 165), DeepSeek-V3.1 139.8 (#113 of 165) and GPT OSS 120B 139.8 (#114 of 165). The confidence ranges of the top two overlap (136.3–141.6 vs 135.6–143.2), so treat the gap as small.
Which is better for coding?
There are no published SWE-bench Verified results for DeepSeek-V3.1, Kimi K2 Instruct and GPT OSS 120B yet, so there is no like-for-like coding score. On overall capability, Kimi K2 Instruct 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?
DeepSeek-V3.1, Kimi K2 Instruct and GPT OSS 120B share the same 131,072-token context window. Maximum output per response: DeepSeek-V3.1 up to 8,192, Kimi K2 Instruct up to 32,768, GPT OSS 120B up to 32,768 tokens.
Which can read images, PDFs, audio or video?
DeepSeek-V3.1 accepts text; Kimi K2 Instruct accepts text; GPT OSS 120B accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights (MIT License and Modified MIT), so you can self-host them.
Which is newer?
DeepSeek-V3.1 is the newest, released Aug 21, 2025. GPT OSS 120B came out Aug 5, 2025; Kimi K2 Instruct came out Jul 11, 2025. Knowledge cutoff: GPT OSS 120B Jun 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.
Related comparisons
- DeepSeek-V3.1 vs Qwen3 235B-A22B Instruct 2507Alibaba (Qwen) · ECI 138.8
- DeepSeek-V3.1 vs Gemini 2.5 Flash Preview 05-20Google · ECI 141.5
- Kimi K2 Instruct vs Gemini 2.5 FlashGoogle · ECI 140.6
- Kimi K2 Instruct vs DeepSeek R1 0528DeepSeek · ECI 141.2
- Kimi K2 Instruct vs Qwen3 235B-A22B Thinking 2507Alibaba (Qwen) · ECI 143.8
- Kimi K2 Instruct vs GLM-4.7Z.ai (Zhipu) · ECI 143.7
