o4-mini-deep-research vs Kimi K2 Instruct 0905 vs Jamba Large
Too close to call on our weighted score (Kimi K2 Instruct 0905 38, o4-mini-deep-research 37, Jamba Large 29). The right pick depends on what you value most.
OpenAIo4-mini-deep-research
37/100- ECI—
- Price$1.80 / $7.20
- Context200K
Moonshot AIKimi K2 Instruct 0905
38/100- ECI—
- Price$0.60 / $2.50
- Context262K
AI21 LabsJamba Large
29/100- ECI—
- Price$2.00 / $8.00
- Context256K
Too close to call
It is close. Our weighted score puts them within 1 points (Kimi K2 Instruct 0905 38/100, o4-mini-deep-research 37/100, Jamba Large 29/100), so choose by what matters most for your work: Kimi K2 Instruct 0905 on price and Kimi K2 Instruct 0905 for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceKimi K2 Instruct 0905Kimi K2 Instruct 0905 $1.07 · o4-mini-deep-research $3.15 · Jamba Large $3.50 per 1M tokens (3:1 blend)
- Longest contextKimi K2 Instruct 0905Kimi K2 Instruct 0905 262,144 · Jamba Large 256,000 · o4-mini-deep-research 200,000 tokens
- Widest inputso4-mini-deep-researcho4-mini-deep-research: Text, Images · Kimi K2 Instruct 0905: Text · Jamba Large: Text
- Self-hostingKimi K2 Instruct 0905 and Jamba LargePublishes downloadable weights (Modified MIT)
| Measure | Weight | o4-mini-deep-research | Kimi K2 Instruct 0905 | Jamba Large |
|---|---|---|---|---|
| Price | 50% | 25 | 46 | 23 |
| Inputs & features | 30% | 60 | 25 | 35 |
| Context window | 20% | 32 | 37 | 36 |
| Overall | 100% | 37/100 | 38/100 | 29/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | o4-mini-deep-researchOpenAI | Kimi K2 Instruct 0905Moonshot AI | Jamba LargeAI21 Labs |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $1.80 | $0.60 (best) | $2.00 |
| Output | $7.20 | $2.50 (best) | $8.00 |
| Cached input | — | — | — |
| Blended (3:1) | $3.15 | $1.07 (best) | $3.50 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 1 providers | Median of 12 providers | Official AI21 Labs API |
| Limits | |||
| Context window | 200,000 tokens | 262,144 tokens (best) | 256,000 tokens |
| Max output | 100,000 tokens (best) | 98,304 tokens | 4,096 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | Yes |
| Availability | |||
| Weights | Proprietary | OpenModified MIT | Open |
| API model ID | — | — | jamba-large |
| API providers | 1 | 13 (best) | 1 |
| Released | Jun 26, 2025 | Sep 5, 2025 | Jul 3, 2025 |
| Knowledge cutoff | May 2024 | — | Aug 22, 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- o4-mini-deep-research$32.40
- Kimi K2 Instruct 0905$11.00
- Jamba Large$36.00
Which should you choose?
Which is better: o4-mini-deep-research, Kimi K2 Instruct 0905 or Jamba Large?
It is close. Our weighted score puts them within 1 points (Kimi K2 Instruct 0905 38/100, o4-mini-deep-research 37/100, Jamba Large 29/100), so choose by what matters most for your work: Kimi K2 Instruct 0905 on price and Kimi K2 Instruct 0905 for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, o4-mini-deep-research, Kimi K2 Instruct 0905 or Jamba Large?
Kimi K2 Instruct 0905 is cheaper at $0.60 input / $2.50 output per million tokens (median across 12 API providers). o4-mini-deep-research costs $1.80 input / $7.20 output per million tokens (median across 1 API provider); Jamba Large costs $2.00 input / $8.00 output per million tokens (official AI21 Labs API price). At a typical mix of three input tokens to one output token, that is $1.07 per million tokens for Kimi K2 Instruct 0905 versus $3.15 for o4-mini-deep-research (2.9× as much) and $3.50 for Jamba Large (3.3× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. o4-mini-deep-research has not been scored yet, Kimi K2 Instruct 0905 has not been scored yet and Jamba Large has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for o4-mini-deep-research, Kimi K2 Instruct 0905 and Jamba Large yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Kimi K2 Instruct 0905 has the largest context window at 262,144 tokens, against 256,000 for Jamba Large and 200,000 for o4-mini-deep-research. Maximum output per response: o4-mini-deep-research up to 100,000, Kimi K2 Instruct 0905 up to 98,304, Jamba Large up to 4,096 tokens.
Which can read images, PDFs, audio or video?
o4-mini-deep-research accepts text and images; Kimi K2 Instruct 0905 accepts text; Jamba Large accepts text. o4-mini-deep-research handles the widest range of inputs.
Are any of these open source?
Kimi K2 Instruct 0905 and Jamba Large publishes its weights (Modified MIT) and can be self-hosted; o4-mini-deep-research is proprietary.
Which is newer?
Kimi K2 Instruct 0905 is the newest, released Sep 5, 2025. Jamba Large came out Jul 3, 2025; o4-mini-deep-research came out Jun 26, 2025. Knowledge cutoff: o4-mini-deep-research May 2024, Jamba Large Aug 22, 2024.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 2, 2026.
