Open Gemma instruction model for efficient chat and self-hosted deployments
Benchmarks
| Benchmark | Category | Score | Variant | Provenance | Source | Observed |
|---|---|---|---|---|---|---|
| AA Agentic Index | agentic | 11.0 | — | mirror | openrouter | Jul 23, 2026 |
| AA Coding Index | coding | 39.3 | — | mirror | openrouter | Jul 23, 2026 |
| AA Intelligence Index | intelligence | 25.7 | — | mirror | openrouter | Jul 23, 2026 |
Includes index data from Artificial Analysis.
Providers
Live operational stats per hosting endpoint, via OpenRouter.
| Provider | Quant | Context | $/1M in · out | Uptime 24h | Latency | Throughput |
|---|---|---|---|---|---|---|
| Cloudflare | — | 256K | $0.1 · $0.3 | 99.04% | — | — |
| DeepInfra | fp8 | 262K | $0.07 · $0.34 | 99.39% | — | — |
| DekaLLM | bf16 | 262K | $0.06 · $0.33 | 98.48% | — | — |
| — | 262K | $0.15 · $0.6 | 74.98% | — | — | |
| NextBit | bf16 | 262K | $0.12 · $0.35 | 99.58% | — | — |
| Novita | bf16 | 262K | $0.13 · $0.4 | 98.87% | — | — |
| Parasail | bf16 | 262K | $0.13 · $0.4 | 99.09% | — | — |
| SiliconFlow | fp8 | 262K | $0.12 · $0.4 | 95.52% | — | — |
| Venice | bf16 | 256K | $0.13 · $0.4 | 98.76% | — | — |
| Wafer | fp8 | 262K | $0.13 · $0.4 | 99.90% | — | — |