Qwen3 30B-A3B Instruct 2507 vs GPT OSS 20B vs Gemini 2.5 Flash-Lite Preview 06-17
Gemini 2.5 Flash-Lite Preview 06-17 comes out ahead, 67 to 61 and 59 on our weighted score, though GPT OSS 20B is 45% cheaper per token.
Alibaba (Qwen)Qwen3 30B-A3B Instruct 2507
59/100- ECI137.3
- Price$0.10 / $0.30
- Context262K
OpenAIGPT OSS 20B
61/100- ECI137.8
- Price$0.07 / $0.25
- Context131K
- Our pick
GoogleGemini 2.5 Flash-Lite Preview 06-17
67/100- ECI133.8
- Price$0.12 / $0.48
- Context1.05M
Gemini 2.5 Flash-Lite Preview 06-17 is our pick
Gemini 2.5 Flash-Lite Preview 06-17 is the better all-round choice, scoring 67/100 against GPT OSS 20B (61) and Qwen3 30B-A3B Instruct 2507 (59). It leads on inputs & features and context window. GPT OSS 20B wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityGPT OSS 20BCapabilities Index (ECI): GPT OSS 20B 137.8 · Qwen3 30B-A3B Instruct 2507 137.3 · Gemini 2.5 Flash-Lite Preview 06-17 133.8
- Lowest priceGPT OSS 20BGPT OSS 20B $0.115 · Qwen3 30B-A3B Instruct 2507 $0.15 · Gemini 2.5 Flash-Lite Preview 06-17 $0.21 per 1M tokens (3:1 blend)
- Longest contextGemini 2.5 Flash-Lite Preview 06-17Gemini 2.5 Flash-Lite Preview 06-17 1,048,576 · Qwen3 30B-A3B Instruct 2507 262,144 · GPT OSS 20B 131,072 tokens
- Widest inputsGemini 2.5 Flash-Lite Preview 06-17Qwen3 30B-A3B Instruct 2507: Text · GPT OSS 20B: Text · Gemini 2.5 Flash-Lite Preview 06-17: Text, Images, PDFs, Audio, Video
- Self-hostingQwen3 30B-A3B Instruct 2507 and GPT OSS 20BPublishes downloadable weights (Apache 2.0)
| Measure | Weight | Qwen3 30B-A3B Instruct 2507 | GPT OSS 20B | Gemini 2.5 Flash-Lite Preview 06-17 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 61 | 61 | 57 |
| Price | 25% | 85 | 90 | 79 |
| Inputs & features | 15% | 25 | 35 | 90 |
| Context window | 10% | 37 | 25 | 61 |
| Overall | 100% | 59/100 | 61/100 | 67/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Qwen3 30B-A3B Instruct 2507Alibaba (Qwen) | GPT OSS 20BOpenAI | Gemini 2.5 Flash-Lite Preview 06-17Google |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 137.3 | 137.8 (best) | 133.8 |
| ECI rank | #124 of 165 | #122 of 165 (best) | #132 of 165 |
| GPQA DiamondGraduate-level science questions | 55.6% | 60.8% (best) | — |
| OTIS Mock AIME 2024–2025Competition mathematics | 62.2% | 65.3% (best) | — |
| Price per million tokens | |||
| Input | $0.10 | $0.07 (best) | $0.12 |
| Output | $0.30 | $0.25 (best) | $0.48 |
| Cached input | — | — | — |
| Blended (3:1) | $0.15 | $0.115 (best) | $0.21 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 12 providers | Median of 37 providers | Median of 2 providers |
| Limits | |||
| Context window | 262,144 tokens | 131,072 tokens | 1,048,576 tokens (best) |
| Max output | 16,384 tokens | 32,768 tokens | 65,536 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | Yes |
| PDFs | No | No | Yes |
| Audio | No | No | Yes |
| Video | No | No | Yes |
| Reasoning | No | Yes | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | OpenApache 2.0 | Open | Proprietary |
| API model ID | — | — | — |
| API providers | 14 | 41 (best) | 2 |
| Released | Jul 28, 2025 | Aug 5, 2025 | Jun 17, 2025 |
| Knowledge cutoff | — | Jun 2024 | Jan 2025 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Qwen3 30B-A3B Instruct 2507$1.60
- GPT OSS 20B$1.20
- Gemini 2.5 Flash-Lite Preview 06-17$2.16
Which should you choose?
Which is better: Qwen3 30B-A3B Instruct 2507, GPT OSS 20B or Gemini 2.5 Flash-Lite Preview 06-17?
Gemini 2.5 Flash-Lite Preview 06-17 is the better all-round choice, scoring 67/100 against GPT OSS 20B (61) and Qwen3 30B-A3B Instruct 2507 (59). It leads on inputs & features and context window. GPT OSS 20B wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, Qwen3 30B-A3B Instruct 2507, GPT OSS 20B or Gemini 2.5 Flash-Lite Preview 06-17?
GPT OSS 20B is cheaper at $0.07 input / $0.25 output per million tokens (median across 37 API providers). Qwen3 30B-A3B Instruct 2507 costs $0.10 input / $0.30 output per million tokens (median across 12 API providers); Gemini 2.5 Flash-Lite Preview 06-17 costs $0.12 input / $0.48 output per million tokens (median across 2 API providers). At a typical mix of three input tokens to one output token, that is $0.115 per million tokens for GPT OSS 20B versus $0.15 for Qwen3 30B-A3B Instruct 2507 (1.3× as much) and $0.21 for Gemini 2.5 Flash-Lite Preview 06-17 (1.8× as much).
Which scores higher on benchmarks?
GPT OSS 20B scores higher on the Capabilities Index (ECI): GPT OSS 20B 137.8 (#122 of 165), Qwen3 30B-A3B Instruct 2507 137.3 (#124 of 165) and Gemini 2.5 Flash-Lite Preview 06-17 133.8 (#132 of 165). The confidence ranges of the top two overlap (133.1–139.7 vs 131.7–139.4), so treat the gap as small.
Which is better for coding?
There are no published SWE-bench Verified results for Qwen3 30B-A3B Instruct 2507, GPT OSS 20B and Gemini 2.5 Flash-Lite Preview 06-17 yet, so there is no like-for-like coding score. On overall capability, GPT OSS 20B 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 2.5 Flash-Lite Preview 06-17 has the largest context window at 1,048,576 tokens, against 262,144 for Qwen3 30B-A3B Instruct 2507 and 131,072 for GPT OSS 20B. Maximum output per response: Qwen3 30B-A3B Instruct 2507 up to 16,384, GPT OSS 20B up to 32,768, Gemini 2.5 Flash-Lite Preview 06-17 up to 65,536 tokens.
Which can read images, PDFs, audio or video?
Qwen3 30B-A3B Instruct 2507 accepts text; GPT OSS 20B accepts text; Gemini 2.5 Flash-Lite Preview 06-17 accepts text, images, PDFs, audio and video. Gemini 2.5 Flash-Lite Preview 06-17 handles the widest range of inputs.
Are any of these open source?
Qwen3 30B-A3B Instruct 2507 and GPT OSS 20B publishes its weights (Apache 2.0) and can be self-hosted; Gemini 2.5 Flash-Lite Preview 06-17 is proprietary.
Which is newer?
GPT OSS 20B is the newest, released Aug 5, 2025. Qwen3 30B-A3B Instruct 2507 came out Jul 28, 2025; Gemini 2.5 Flash-Lite Preview 06-17 came out Jun 17, 2025. Knowledge cutoff: GPT OSS 20B Jun 2024, Gemini 2.5 Flash-Lite Preview 06-17 Jan 2025.
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
- Qwen3 30B-A3B Instruct 2507 vs DeepSeek-V3.1DeepSeek · ECI 139.8
- Qwen3 30B-A3B Instruct 2507 vs Llama 4 Maverick 17B InstructMeta · ECI 132.3
- GPT OSS 20B vs DeepSeek-V3.1DeepSeek · ECI 139.8
- GPT OSS 20B vs Kimi K2 InstructMoonshot AI · ECI 139.9
- Gemini 2.5 Flash-Lite Preview 06-17 vs Qwen3 30B A3BAlibaba (Qwen) · ECI 136.2
- Gemini 2.5 Flash-Lite Preview 06-17 vs Llama 4 Maverick 17B InstructMeta · ECI 132.3
