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 | gpt-5.5-pre-release | Z.ai: GLM 5.2 |
|---|---|---|
| Known Good Index | 98 | 91 |
| Mathematics Index | 100 | 86 |
| Agentic Index | 94 | 91 |
| Openness Index | 5 | 60 |
Quality Breakdown
Every evaluation both models report a published score for
Bar = gpt-5.5-pre-release · marker = Z.ai: GLM 5.2 · scores ingested from Epoch AI (CC-BY 4.0)
Right-hand figures read gpt-5.5-pre-release / Z.ai: GLM 5.2. gpt-5.5-pre-release leads on 3 of 3, Z.ai: GLM 5.2 on 0.
Arena Elo
Human preference rating from blind pairwise votes
Elo from blind pairwise votes, ingested from Arena. Higher is better.
| Metric | gpt-5.5-pre-release | Z.ai: GLM 5.2 |
|---|---|---|
| Text Arena | — | 1,464 |
| Text Arena (style controlled) | — | 1,469 |
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 | gpt-5.5-pre-release | Z.ai: GLM 5.2 |
|---|---|---|
| Input $/1M | — | $0.76 |
| Output $/1M | — | $2.38 |
| Blended $/1M | — | $1.16 |
| Cache read $/1M | — | $0.141 |
Context Window
Maximum input accepted and maximum output emitted
Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.
| Metric | gpt-5.5-pre-release | Z.ai: GLM 5.2 |
|---|---|---|
| Context window | — | 1.05M |
| Max output tokens | — | 131k |
Speed & Latency
Median throughput and time to first token across hosting providers
Medians across all tracked providers, derived from OpenRouter endpoint telemetry.
| Metric | gpt-5.5-pre-release | Z.ai: GLM 5.2 |
|---|---|---|
| Output tokens/s | — | 48 |
| Time to first token | — | 2.2s |
Providers
Which hosts serve each model, and at what price and speed
| Provider | gpt-5.5-pre-release blended | gpt-5.5-pre-release tok/s | Z.ai: GLM 5.2 blended | Z.ai: GLM 5.2 tok/s |
|---|---|---|---|---|
| BaseTen | — | — | $2.15 | 150 |
| Friendli | — | — | $2.15 | 101 |
| Together | — | — | $2.15 | 98 |
| Wafer | — | — | $1.94 | 91 |
| Ionstream | — | — | $2.15 | 83 |
| Fireworks | — | — | $2.15 | 80 |
| Decart | — | — | $1.52 | 77 |
| Chutes | — | — | $1.93 | 68 |
| CoreWeave | — | — | $1.18 | 67 |
| Alibaba · first party | — | — | $1.24 | 65 |
| Parasail | — | — | $2.15 | 65 |
| DeepInfra | — | — | $1.45 | 56 |
| Io Net | — | — | $4.32 | 54 |
| Morph | — | — | $1.85 | 54 |
| Baidu · first party | — | — | $1.07 | 53 |
| Cloudflare | — | — | $2.15 | 53 |
| Phala | — | — | $2.15 | 50 |
| Sail Research | — | — | $1.62 | 47 |
| AkashML | — | — | $1.18 | 44 |
| Ambient | — | — | $1.89 | 44 |
| GMICloud | — | — | $1.42 | 37 |
| AtlasCloud | — | — | $1.94 | 35 |
| Venice | — | — | $2.15 | 35 |
| Z.AI · first party | — | — | $2.15 | 32 |
| SiliconFlow | — | — | $2.00 | 31 |
| Novita | — | — | $1.16 | 27 |
| DigitalOcean | — | — | $1.89 | 26 |
| StreamLake | — | — | $1.16 | 26 |
| Inceptron | — | — | $1.43 | 8 |
Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.