Quality
Known Good Index and the capability indices, side by side
Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.
| Metric | DeepSeek: DeepSeek V3 | gpt-4.5-preview |
|---|---|---|
| Known Good Index | 44 | 47 |
| Mathematics Index | 40 | 59 |
| Science Index | 53 | 68 |
| Reasoning Index | 52 | 48 |
| Openness Index | 60 | 5 |
Quality Breakdown
Every evaluation both models report a published score for
Bar = DeepSeek: DeepSeek V3 · marker = gpt-4.5-preview · scores ingested from Epoch AI (CC-BY 4.0), LiveBench (Apache-2.0)
Right-hand figures read DeepSeek: DeepSeek V3 / gpt-4.5-preview. DeepSeek: DeepSeek V3 leads on 1 of 10, gpt-4.5-preview on 9.
Arena Elo
Human preference rating from blind pairwise votes
Elo from blind pairwise votes, ingested from Arena. Higher is better.
| Metric | DeepSeek: DeepSeek V3 | gpt-4.5-preview |
|---|---|---|
| Text Arena | 1,333 | — |
| Text Arena (style controlled) | 1,358 | — |
Price & Cost
Published API pricing by token type
USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.
| Metric | DeepSeek: DeepSeek V3 | gpt-4.5-preview |
|---|---|---|
| Input $/1M | $0.20 | — |
| Output $/1M | $0.80 | — |
| Blended $/1M | $0.35 | — |
| Cache read $/1M | $0.003 | — |
Context Window
Maximum input accepted and maximum output emitted
Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.
| Metric | DeepSeek: DeepSeek V3 | gpt-4.5-preview |
|---|---|---|
| Context window | 164k | — |
| Max output tokens | 16k | — |
Speed & Latency
Median throughput and time to first token across hosting providers
Medians across all tracked providers, derived from OpenRouter endpoint telemetry.
| Metric | DeepSeek: DeepSeek V3 | gpt-4.5-preview |
|---|---|---|
| Output tokens/s | 21 | — |
| Time to first token | 1.1s | — |
Providers
Which hosts serve each model, and at what price and speed
| Provider | DeepSeek: DeepSeek V3 blended | DeepSeek: DeepSeek V3 tok/s | gpt-4.5-preview blended | gpt-4.5-preview tok/s |
|---|---|---|---|---|
| StreamLake | $0.35 | 24 | — | — |
| DeepInfra | $0.46 | 20 | — | — |
| Novita | $0.62 | 14 | — | — |
Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.