LLM Comparison
North Mini Code vs Reka Flash 3
Side-by-side specs, pricing & capabilities
Add to comparison
2/6 models| Organization | ||
| OpenTools Score | ||
| Family | North | Reka |
| Status | Current | Current |
| Release Date | Jun 2026 | Mar 2025 |
| Context Window | 262K tokens | 66K tokens |
| Input Price | Free | $0.10/M tokens |
| Output Price | Free | $0.20/M tokens |
| Pricing Notes | Cohere's official model documentation states North Mini Code is free for trial and production API keys until rate limits are reached; Model Vault production deployment is also supported. | — |
| Capabilities | reasoningagentic-codingtool-usestructured-outputmultilinguallocal-deployment | textcode |
| Max Output | 66K tokens | 66K tokens |
| API Identifier | north-mini-code-1-0 | rekaai/reka-flash-3 |
| Benchmarks | ||
| AIME 2025 | — | 33.7%artificial-analysis |
| GPQA Diamond | 75.7%artificial-analysis | 52.9%artificial-analysis |
| SciCode | 38.8%artificial-analysis | — |
| Terminal-Bench 2.1 | 35.6%artificial-analysis | — |
| View North Mini Code | View Reka Flash 3 | |
Cost Calculator
Enter your expected monthly token usage to compare costs.
| Model | Input | Output | Total / mo | vs Best |
|---|---|---|---|---|
| North Mini CodeCheapest | $0.00 | $0.00 | $0.00 | — |
| Reka Flash 3 | $0.10 | $0.10 | $0.20 | +0% |
Compare hosting configurations
Compare two exact setups, including different configurations of the same model. Hardware fit, measured speed and costs are separate decisions.
Add your cost assumptions (optional)
Use hours for the same scenario period. Cloud billing includes idle time; active hours alone do not set the bill. Local electricity uses whole-system attributable power, powered hours and your tariff. Missing inputs remain unknown.
Measured speed needs the same pinned workload protocol, tokenizer, context, batching and cache state.
| Configuration | Run complete native North Mini Code with retained reasoning and inert tools on a CPU workstation | Run Reka Flash 3 bounded reasoning on a 32 GiB Intel desktop |
|---|---|---|
| Model | North Mini Code | Reka Flash 3 |
| Hardware | Intel Xeon w5-2465X / ASUS W790E-SAGE SE,256GiB RAM,400GiB free disk | Intel Core i5-12400 desktop, 4 inference threads, 32 GiB RAM |
| Runtime | Complete native North Mini Code CPU with Command4 parsing Transformers 5.11.0; Torch 2.11.0 CPU; Cohere Melody 0.15.2; Python 3.12 · CPU | llama.cpp v0.7.0 (620d2225cc436ab0b001e3200e590e3b286ef9ab) · CPU |
| Artifact | North Mini Code 1.0 complete native 49-part Cohere2Moe checkpoint · Full native BF16 checkpoint loaded as CPU FP32; all 128 experts retained, not active-parameter storage · Unquantized native BF16 weights; reader explicitly loads the complete model as CPU FP32 | Reka Flash 3 Q4_K_M, Unsloth derivative · GGUF Q4_K_M mixed quantization · Q4_K_M mixed quantization published by Unsloth |
| Context / concurrency | 4,096 tokens / 1 | 8,192 tokens / 1 |
| Required memory | Unknown (unknown) | Unknown (unknown) |
| Evidence | Documented setup | Documented setup |
| Hardware fit | Hardware fit unknown Complete FP32 load, cache, expert routing, temporary tensors, environment and OS peak are unmeasured. Enter available memory per device or unified memory pool. Host RAM: 256 GiB is a selected host allocation, not a measured process minimum or fit result. Free disk space: 400 GiB free/root allocation is a reader planning choice, not a measured minimum. | Hardware fit unknown No exact CPU process peak for this 21B dense GGUF, 8K context and native reasoning workload. Enter available memory per device or unified memory pool. Host RAM: 32 GiB is chosen installed host capacity, not a measured minimum. Free disk space: 40 GiB free/root disk is a walkthrough allocation, not measured minimum. |
| One-time cost | Unknown hardware: Existing desktop purchase price and incremental hardware cost depend on the reader; unknown. setup: Initial provisioning/build time and labor cost are unknown. | Unknown hardware: Existing desktop purchase price and incremental hardware cost depend on the reader; unknown. setup: Initial provisioning/build time and labor cost are unknown. |
| Recurring scenario cost | Unknown electricity: Actual wall power, electricity rate and powered hours are unknown; no TDP-to-wall-power conversion. storage: Incremental cost of local disk capacity is unknown. The 400 GiB free-disk walkthrough allocation is an author choice, not a measured requirement. egress: Data transfer price, allowances and traffic volume are unknown. administration: Maintenance labor rate and hours are unknown. One-time setup and recurring costs are separate. Unknown components are not treated as free. No monthly token output is inferred from peak throughput. | Unknown electricity: Actual wall power, electricity rate and powered hours are unknown; no TDP-to-wall-power conversion. storage: Incremental cost of local disk capacity is unknown. The 40 GiB free-disk walkthrough allocation is an author choice, not a measured requirement. egress: Data transfer price, allowances and traffic volume are unknown. administration: Maintenance labor rate and hours are unknown. One-time setup and recurring costs are separate. Unknown components are not treated as free. No monthly token output is inferred from peak throughput. |
| Published measurements (unranked) | No compatible measurement available | No compatible measurement available |
Different currencies and incomplete costs are not ranked. Model quality benchmarks remain on the model pages; they do not establish quality or speed for every deployment.
Cohere
North Mini Code
North Mini Code is Cohere's open-weight Mixture-of-Experts model for agentic coding, repository-level software engineering, terminal workflows, tool use, and local deployment.
Reka
Reka Flash 3
Reka Flash 3 is a large language model from Reka. Supports up to 65,536 token context window. Available from $0.10/M input tokens.
More Comparisons
Looking for more AI models?
Browse All LLMs