Open Gemma instruction model for efficient chat and self-hosted deployments
Benchmarks
| Benchmark | Category | Score | Variant | Provenance | Source | Observed |
|---|---|---|---|---|---|---|
| AA Agentic Index | agentic | 2.9 | — | mirror | openrouter | Oct 5, 2026 |
| AA Coding Index | coding | 39.3 | — | mirror | openrouter | Jul 23, 2026 |
| AA Intelligence Index | intelligence | 16.7 | — | mirror | openrouter | Oct 5, 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 |
|---|---|---|---|---|---|---|
| AkashML | fp4 | 131K | $0.07 · $0.24 | 99.93% | — | — |
| Cloudflare | — | 256K | $0.1 · $0.3 | 99.65% | — | — |
| CoreWeave | bf16 | 262K | $0.1 · $0.3 | 99.24% | — | — |
| Darkbloom | — | 131K | $0.04 · $0.15 | 100.00% | — | — |
| DeepInfra | fp8 | 262K | $0.07 · $0.34 | 99.79% | — | — |
| DekaLLM | bf16 | 262K | $0.06 · $0.33 | 99.93% | — | — |
| — | 262K | $0.15 · $0.6 | 63.19% | — | — | |
| Io Net | bf16 | 262K | $0.067 · $0.29 | 99.93% | — | — |
| Makora | — | 256K | $0.09 · $0.28 | 100.00% | — | — |
| NextBit | bf16 | 262K | $0.076 · $0.255 | 100.00% | — | — |
| Novita | bf16 | 262K | $0.13 · $0.4 | 99.93% | — | — |
| Parasail | bf16 | 262K | $0.13 · $0.4 | 100.00% | — | — |
| PrimeIntellect | bf16 | 262K | $0.1 · $0.3 | 100.00% | — | — |
| SiliconFlow | fp8 | 262K | $0.14 · $0.4 | 99.02% | — | — |
| Venice | bf16 | 256K | $0.13 · $0.4 | 100.00% | — | — |