Jamba 1.7
LLM Provider / Model · Long-context RAG with low memory; cost-conscious enterprise deployments
At a glance
| Input price | $0.40 |
| Output price | $1.20 |
| Cache price | $0.05 |
| Price tier | Budget |
| Context window | 256K |
| Max output | 16K |
| Context tier | 128K-500K |
| Speed tier | Fast |
| Latency | low |
| Knowledge cutoff | Aug 2025 |
| Modality | Text |
| Model ID | jamba-1.7-large |
| Provider | AI21 Labs |
| HumanEval | 86% |
| MMLU | 81% |
| SWE-Bench | 52% |
| Benchmark | Top throughput for long-context RAG at low memory cost |
| Released | Oct 2025 |
| Hosting | Closed/API + Self-host |
| Capabilities | Tool use, Hybrid SSM-Transformer, Function calling, Long-context efficient, RAG-tuned |
What Jamba 1.7 does
Tool use, Hybrid SSM-Transformer, Function calling, Long-context efficient, RAG-tuned
Best for
Long-context RAG with low memory; cost-conscious enterprise deployments
Works well with
Agent / Orchestration
Conflicts & caveats
- Trae's built-in model menu may not include Jamba 1.7 — Trae primarily ships with Claude/GPT/Gemini/DeepSeek.
- Google Antigravity is built around Gemini 3 Pro; Claude and recent GPT-5 models also work, but Jamba 1.7 is not in the supported model list.
- Kiro's spec mode is tuned for Claude models — Jamba 1.7 may work but spec-driven flows are best with Claude.
- Cursor's built-in model menu may not include Jamba 1.7 — verify it's available in your Cursor model picker.
- OpenAI Codex (VS Code) is tied to OpenAI's GPT-5.1 Codex family — Jamba 1.7 is not selectable inside this integration.
- Gemini CLI is locked to Google Gemini models — your LLM is Jamba 1.7.
- Privacy conflict: Self-hosted Jamba 1.7 sends code to cloud @codebase (Cursor). Use local context (Continue indexing, ChromaDB, LanceDB, pgvector, Vespa self-hosted) for true privacy.
- Claude Code only works with Claude models — your LLM is Jamba 1.7.
- Devin uses Cognition's proprietary model internally — your selected Jamba 1.7 will be ignored.
- OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected Jamba 1.7 will be ignored.
- Jules runs Gemini models internally — your selected Jamba 1.7 will be ignored.
- The Claude Code SDK is tuned for Claude models — Jamba 1.7 may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
- Lovable uses a fixed internal model — your selected Jamba 1.7 will be ignored.
- Replit Agent 3 uses its own model selection internally — your selected Jamba 1.7 will be ignored.
Build a full stack around Jamba 1.7 — Flowpicker shows compatibility warnings before you commit.
Open the stack planner →