Phi-4-mini vs Qwen2.5-Coder-0.5B vs Reka Flash 3
No independent capability benchmark covers every model here yet. The score uses context window.
Microsoft
Phi-4-mini
32/100- ECI—
- Price$0.075 / $0.30
- Context128K
Alibaba (Qwen)
Qwen2.5-Coder-0.5B
11/100- ECI68.1
- Price—
- Context33K
Reka AI
Reka Flash 3
22/100- ECI—
- Price$0.10 / $0.20
- Context66K
Microsoft
Phi-4-mini
32/100- ECI—
- Price$0.075 / $0.30
- Context128K
Alibaba (Qwen)
Qwen2.5-Coder-0.5B
11/100- ECI68.1
- Price—
- Context33K
Reka AI
Reka Flash 3
22/100- ECI—
- Price$0.10 / $0.20
- Context66K
Capability evidence is incomplete
There is not enough shared capability evidence to name an overall winner. The scores (Phi-4-mini 32/100, Reka Flash 3 22/100, Qwen2.5-Coder-0.5B 11/100) compare the available specifications and price. The score weighs context window 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.There is not enough shared capability evidence to name an overall winner. The scores (Phi-4-mini 32/100, Reka Flash 3 22/100, Qwen2.5-Coder-0.5B 11/100) compare the available specifications only. The score weighs context window 100%. 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 priceReka Flash 3Reka Flash 3 $0.125 · Phi-4-mini $0.131 per 1M tokens (3:1 blend) · Qwen2.5-Coder-0.5B unpriced
- Longest contextPhi-4-miniPhi-4-mini 128,000 · Reka Flash 3 65,536 · Qwen2.5-Coder-0.5B 32,768 tokens
- Widest inputsSame inputsPhi-4-mini: Text · Qwen2.5-Coder-0.5B: Text · Reka Flash 3: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Phi-4-mini | Qwen2.5-Coder-0.5B | Reka Flash 3 |
|---|---|---|---|---|
| Context window | 100% | 32 | 11 | 22 |
| Specs & price only | 100% | 32/100 | 11/100 | 22/100 |
Left out because at least one model lacks the data: capability, price and inputs & features. The remaining weights were rescaled.
| Measure | Weight | Phi-4-mini | Qwen2.5-Coder-0.5B | Reka Flash 3 |
|---|---|---|---|---|
| Context window | 100% | 32 | 11 | 22 |
| Specs only | 100% | 32/100 | 11/100 | 22/100 |
Left out because at least one model lacks the data: capability and inputs & features. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Unavailable specifications are marked “Not published”. Publisher results include their evaluation settings and source.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | Not published | 68.1 | Not published |
| ECI rank | Not published | #159 of 159 | Not published |
| ECI confidence interval | Not published | 48.4–83.9 | Not published |
| ECI evaluation date | Not published | Sep 18, 2024 | Not published |
| Documented capabilities | Compact Microsoft instruction model tuned for efficient coding assistance, reasoning, and low-latency agent tasks | Tiny open Qwen code model for lightweight completion and on-device coding | Efficient model for low-latency assistance, extraction, and routine automation |
| Model parameters | Not published | Not published | Not published |
| Output modalities | Text | Text | Text |
| Publisher-reported benchmarks | |||
| MMLUPublisher-reported result | 67.3accuracyReported result | Not published | Not published |
| Price per million tokens | |||
| Input | $0.075 (best) | Not published | $0.10 |
| Output | $0.30 | Not published | $0.20 (best) |
| Cached input | Not published | Not published | Not published |
| Blended (3:1) | $0.131 | Not published | $0.125 (best) |
| Long-context rate | Same rate | Not published | Same rate |
| Price source | Official Azure API | Not published | Median of 3 providers |
| Limits | |||
| Context window | 128,000 tokens (best) | 32,768 tokens | 65,536 tokens |
| Max input | Not published | Not published | Not published |
| Max output | 4,096 tokens | 8,192 tokens | 58,982 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 | No | No | Yes |
| Tool calling | Yes | No | No |
| Structured output | Not published | Not published | Not published |
| File attachments | No | No | No |
| Temperature control | Yes | Yes | Yes |
| Reasoning control | Not published | Not published | always |
| Availability | |||
| Weights | Open | OpenApache 2.0 | Open |
| API model ID | phi-4-mini | Not published | Not published |
| API providers | 2 | Not published | 3 (best) |
| Released | Feb 26, 2025 | Nov 12, 2024 | Mar 10, 2025 |
| Knowledge cutoff | Jun 2024 | Not published | Jan 31, 2025 |
| Publisher documentation | Publisher documentation · Model weights | Model weights | Model weights |
| Source metadata updated | Feb 26, 2025 | Nov 12, 2024 | Mar 10, 2025 |
| Data sources | models.dev · Epoch AI | models.dev · Epoch AI | models.dev · Epoch AI |
| Catalog checked | Nov 22, 2026 | Nov 22, 2026 | Nov 22, 2026 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
Phi-4-mini$1.35
Qwen2.5-Coder-0.5B—
Reka Flash 3$1.40
Which should you choose?
Which is better: Phi-4-mini, Qwen2.5-Coder-0.5B or Reka Flash 3?
There is not enough shared capability evidence to name an overall winner. The scores (Phi-4-mini 32/100, Reka Flash 3 22/100, Qwen2.5-Coder-0.5B 11/100) compare the available specifications and price. The score weighs context window 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.There is not enough shared capability evidence to name an overall winner. The scores (Phi-4-mini 32/100, Reka Flash 3 22/100, Qwen2.5-Coder-0.5B 11/100) compare the available specifications only. The score weighs context window 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Phi-4-mini, Qwen2.5-Coder-0.5B or Reka Flash 3?
Reka Flash 3 is cheaper at $0.10 input / $0.20 output per million tokens (median across 3 API providers). Phi-4-mini costs $0.075 input / $0.30 output per million tokens (official Azure API price). At a typical mix of three input tokens to one output token, that is $0.125 per million tokens for Reka Flash 3 versus $0.131 for Phi-4-mini (1× as much). Qwen2.5-Coder-0.5B has no published per-token price.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Phi-4-mini has not been scored yet, Qwen2.5-Coder-0.5B has an ECI of 68.1 and Reka Flash 3 has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Phi-4-mini, Qwen2.5-Coder-0.5B and Reka Flash 3 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Qwen2.5-Coder-0.5B and Reka Flash 3 has no confirmed tool-calling support in this comparison.
Which has the bigger context window?
Phi-4-mini has the largest context window at 128,000 tokens, against 65,536 for Reka Flash 3 and 32,768 for Qwen2.5-Coder-0.5B. Maximum output per response: Phi-4-mini up to 4,096, Qwen2.5-Coder-0.5B up to 8,192, Reka Flash 3 up to 58,982 tokens.
Which can read images, PDFs, audio or video?
Phi-4-mini accepts text; Qwen2.5-Coder-0.5B accepts text; Reka Flash 3 accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights (Apache 2.0), so you can self-host them.
Which is newer?
Reka Flash 3 is the newest, released Mar 10, 2025. Phi-4-mini came out Feb 26, 2025; Qwen2.5-Coder-0.5B came out Nov 12, 2024. Knowledge cutoff: Phi-4-mini Jun 2024, Reka Flash 3 Jan 31, 2025.
How do you decide the winner?
This rating uses context window (100%, 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 22, 2026.