The AI arena is free today

Open Superagent

GPT OSS 120B vs Kimi K2 0905

Both models are evenly matched across the benchmarks. GPT OSS 120B is 6.0x cheaper per token.

OpenAI · Moonshot AI · Updated for 2026

Which is better?

GPT OSS 120B outperforms in 1 benchmarks (GPQA), while Kimi K2 0905 is better at 1 benchmark (MMLU). Both models are evenly matched across the benchmarks.

On price, GPT OSS 120B is roughly 6.0x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Kimi K2 0905 also accepts a larger context window (262,144 input tokens), making it the stronger choice for long documents and large codebases.

Based on current benchmark, pricing, and model metadata for 2026.

Choose GPT OSS 120B

  • cost matters — it's about 6.0x cheaper per token
  • you need open weights you can self-host or fine-tune

Choose Kimi K2 0905

  • you process long inputs — it offers a 262,144 token context window
  • you want the most recent training data — it shipped Sep 2025

At a glance

The differences that matter most.

Benchmark wins
1 of 2
1 of 2
Input price
$0.09 / M
$0.60 / M
Output price
$0.45 / M
$2.50 / M
Context window
131,072
262,144
Released
Aug 2025
Sep 2025
License
Apache 2.0
Proprietary

Performance Benchmarks

Comparative analysis across standard metrics

2 benchmarks

GPT OSS 120B outperforms in 1 benchmarks (GPQA), while Kimi K2 0905 is better at 1 benchmark (MMLU).

Both models are evenly matched across the benchmarks.

Tue Aug 25 2026 • llm-stats.com

Arena Performance

Playground indexes and blind preference scores

Pricing Analysis

Price comparison per million tokens

GPT OSS 120B costs less

For input processing, GPT OSS 120B ($0.09/1M tokens) is 6.7x cheaper than Kimi K2 0905 ($0.60/1M tokens).

For output processing, GPT OSS 120B ($0.45/1M tokens) is 5.6x cheaper than Kimi K2 0905 ($2.50/1M tokens).

In conclusion, Kimi K2 0905 is more expensive than GPT OSS 120B.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Tue Aug 25 2026 • llm-stats.com
OpenAI
GPT OSS 120B
Input tokens$0.09
Output tokens$0.45
Best providerDeepinfra
Moonshot AI
Kimi K2 0905
Input tokens$0.60
Output tokens$2.50
Best providerNovita
Notice missing or incorrect data?Start an Issue

Model Size

Parameter count comparison

883.2B diff

Kimi K2 0905 has 883.2B more parameters than GPT OSS 120B, making it 756.2% larger.

OpenAI
GPT OSS 120B
116.8Bparameters
Moonshot AI
Kimi K2 0905
1.0Tparameters
116.8B
GPT OSS 120B
1000.0B
Kimi K2 0905

Context Window

Maximum input and output token capacity

Kimi K2 0905 accepts 262,144 input tokens compared to GPT OSS 120B's 131,072 tokens. Kimi K2 0905 can generate longer responses up to 262,144 tokens, while GPT OSS 120B is limited to 131,072 tokens.

OpenAI
GPT OSS 120B
Input131,072 tokens
Output131,072 tokens
Moonshot AI
Kimi K2 0905
Input262,144 tokens
Output262,144 tokens
Tue Aug 25 2026 • llm-stats.com

License

Usage and distribution terms

GPT OSS 120B is licensed under Apache 2.0, while Kimi K2 0905 uses a proprietary license.

License differences may affect how you can use these models in commercial or open-source projects.

GPT OSS 120B

Apache 2.0

Open weights

Kimi K2 0905

Proprietary

Closed source

Release Timeline

When each model was launched

GPT OSS 120B was released on 2025-08-05, while Kimi K2 0905 was released on 2025-09-05.

Kimi K2 0905 is 1 month newer than GPT OSS 120B.

GPT OSS 120B

Aug 5, 2025

1.1 years ago

Kimi K2 0905

Sep 5, 2025

11 months ago

1mo newer

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Provider Availability

GPT OSS 120B is available from DeepInfra, Novita, OpenAI, Fireworks, Groq. Kimi K2 0905 is available from Novita.

GPT OSS 120B

deepinfra logo
Deepinfra
Input Price:Input: $0.09/1MOutput Price:Output: $0.45/1M
novita logo
Novita
Input Price:Input: $0.10/1MOutput Price:Output: $0.50/1M
openai logo
OpenAI
Input Price:Input: $0.10/1MOutput Price:Output: $0.50/1M
fireworks logo
Fireworks
Input Price:Input: $0.15/1MOutput Price:Output: $0.60/1M
groq logo
Groq
Input Price:Input: $0.15/1MOutput Price:Output: $0.60/1M

Kimi K2 0905

novita logo
Novita
Input Price:Input: $0.60/1MOutput Price:Output: $2.50/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against GPT OSS 120B and Kimi K2 0905 side-by-side, then vote on the output you prefer.

GPT OSS 120B
✓ Preferred
Kimi K2 0905
Open in Playground

FAQ

Common questions about GPT OSS 120B vs Kimi K2 0905.

Which is better, GPT OSS 120B or Kimi K2 0905?

Both models are evenly matched across the benchmarks. GPT OSS 120B is made by OpenAI and Kimi K2 0905 is made by Moonshot AI. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does GPT OSS 120B compare to Kimi K2 0905 in benchmarks?

GPT OSS 120B scores MMLU: 90.0%, CodeForces: 82.1%, GPQA: 80.1%, TAU-bench Retail: 67.8%, HealthBench: 57.6%. Kimi K2 0905 scores HumanEval: 94.5%, MMLU: 90.2%, MATH: 89.1%, MMLU-Pro: 82.5%, GPQA: 75.8%.

Is GPT OSS 120B cheaper than Kimi K2 0905?

GPT OSS 120B is 6.7x cheaper for input tokens. GPT OSS 120B costs $0.09/M input and $0.45/M output via deepinfra. Kimi K2 0905 costs $0.60/M input and $2.50/M output via novita.

What are the context window sizes for GPT OSS 120B and Kimi K2 0905?

GPT OSS 120B supports 131K tokens and Kimi K2 0905 supports 262K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between GPT OSS 120B and Kimi K2 0905?

Key differences include context window (131K vs 262K), input pricing ($0.09 vs $0.60/M), licensing (Apache 2.0 vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes GPT OSS 120B and Kimi K2 0905?

GPT OSS 120B is developed by OpenAI and Kimi K2 0905 is developed by Moonshot AI.